{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"id": "4d6d065a-be52-4462-9d33-fd12928aea6a",
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"os.environ['CUDA_VISIBLE_DEVICES'] = '0'"
]
},
{
"cell_type": "markdown",
"id": "036f012a-0623-494b-95f2-35ad8c4a604f",
"metadata": {},
"source": [
"### Tutorial B6: Design\n",
"\n",
"> ersatz (substitute) + predict + ersatz (substitute) + predict + ..."
]
},
{
"cell_type": "markdown",
"id": "1f33fe28-fe87-4f70-948a-8b18cfce5154",
"metadata": {},
"source": [
"Most of the functions we've discussed so far center around extracting what a sequence-based machine learning model has learned after training. Marginalization experiments show the individual effect that each motif has on model predictions; ISM shows the individual effect that each mutation has on model predictions within an observed sequence. However, these models can be used for more than just identifying relevent sequence features and their syntax rules.\n",
"\n",
"For example, trained sequence-based models can be used in the design setting. Although these are numerous methods for designing sequences, each with their own set of important details, the overall approach is to train a model that predicts some sort of readout of interest and then perturb the sequence input until the predicted output matches what is desired by the user. Sometimes, this involves starting with an informative sequence and designing edits, other times this involves the de novo generation of sequences from scratch that exhibit the desired characters. Regardless, the trained model is considered to be an \"oracle\" that outputs how good the in-progress sequence is.\n",
"\n",
"#### Greedy Substitution\n",
"\n",
"The most conceptually simple design method is that of greedy substitution. Given an initial sequence, a predictive model, an output goal, and a set of motifs that one could substitute into the sequence, the task is to find which motifs and what exact positioning should be used to achieved a desired output from the predictive model. Because the aim here is to be conceptually simple, the method proceeds across several iterations where at each iteration, every motif is tried at every position in the sequence, and the motif+position pair that yields the largest improvement in terms of getting the model towards its goal. Naturally, this can be very time consuming but is the most straightforward approach.\n",
"\n",
"Let's start by loading up the Beluga model again."
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "4c78f4e4-4cb3-4013-89d6-a3d72a68ec4c",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import torch\n",
"from model import Beluga\n",
"\n",
"model = Beluga()\n",
"model.load_state_dict(torch.load(\"deepsea.beluga.torch\"))"
]
},
{
"cell_type": "markdown",
"id": "889446fe-2114-44c6-bec4-1aa809b6f6fb",
"metadata": {},
"source": [
"Let's see if we can get this model to design a sequence that predicts high AP-1 binding. Yes, this task is unrealistically simple because AP-1 factors bind to a known motif and one could just insert that into the sequence, but it's a good initial task to show how to use the function. \n",
"\n",
"For the purpose of this example, we can randomly generate a single sequence and use a subset of predefined motifs from JASPAR. Here, we have inserted a motif that we known the Beluga AP-1 tasks respond to as the third motif in the list. "
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "61fd30eb-ff58-4dfe-965a-c40c8016024d",
"metadata": {},
"outputs": [],
"source": [
"import numpy\n",
"from tangermeme.utils import random_one_hot\n",
"\n",
"X = random_one_hot((1, 4, 2000), random_state=0).type(torch.float32)\n",
"motifs = [\n",
" 'GCTAATTAAC',\n",
" 'ATGCCCACC',\n",
" \"GTGACTCATC\",\n",
" 'AGAACAGAATGTTCT',\n",
" 'TGATGACGTCATCGC',\n",
" 'ACATTCCA',\n",
" 'GGGAGGAGGGAGAGGAGGAG',\n",
" 'TAATCGATTA',\n",
" 'CGTCTAGACA',\n",
" 'TGCTATTTTTAG',\n",
" 'CATTGTTTATTT',\n",
" 'TTTCACACCTAGGTGTGAAA',\n",
" 'GGCACGCGCC'\n",
"]"
]
},
{
"cell_type": "markdown",
"id": "521a90a2-a718-43ec-a84c-59a5897741d7",
"metadata": {},
"source": [
"Next, we need a target that we are trying to get the model to predict with the designed sequence. In this case, we probably want all of the tasks for AP-1 factor proteins (e.g., c-/FOS, c-/JUN, etc). Since the predictions from Beluga are in logit space, we can set it to a relatively high positive value to indicate that we want binding of those specific tasks. Note that the first dimension is `1`. Although this method will accept a batch of sequences, the batch size must be `1` because, here, you are designing a single sequence."
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "e5114ea7-b91c-490a-b823-3650ab83d61c",
"metadata": {},
"outputs": [],
"source": [
"names = numpy.loadtxt(\"beluga_target_names.txt\", delimiter=',', dtype=str)\n",
"idxs = torch.tensor(['jun' in n.lower() or 'fos' in n.lower() for n in names])\n",
"\n",
"y = torch.zeros(1, 2002)\n",
"y[:, idxs] = 3"
]
},
{
"cell_type": "markdown",
"id": "ff622d6e-b217-41e9-8705-3840b24de721",
"metadata": {},
"source": [
"Now, we can call the function. This function has a similar signature to other `tangermeme` functions in that it takes in the predictive model, the initial sequence, and a list of motifs. It differs from other functions in that it also takes in `y`, which is the target that we want the predictive model to output. Optionally, `greedy_substitution` can take in a mask that is applied to the predictions from the model and the target. Basically, if your model makes predictions for more outputs than you care about, this mask lets you remove some of those outputs from the loss. In our case, we can remove all predictions not related to the AP-1 factors and say that we want to design a sequence with the highest binding of AP-1 factors regardless of what happens to all other tasks."
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "f830db0f-c013-4962-be98-e6873b6df227",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 0 -- Loss: 72.96, Improvement: N/A, Idx: N/A, Time (s): 0s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 3.05it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 1 -- Loss: 14.17, Improvement: 58.79, Motif Idx: 17, Pos Idx: 966, Time (s): 8.531\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 3.05it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 2 -- Loss: 7.049, Improvement: 7.118, Motif Idx: 15, Pos Idx: 1031, Time (s): 8.525\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 3.02it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 3 -- Loss: 4.952, Improvement: 2.097, Motif Idx: 2, Pos Idx: 1068, Time (s): 8.622\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 3.03it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 4 -- Loss: 4.222, Improvement: 0.7304, Motif Idx: 10, Pos Idx: 984, Time (s): 8.584\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.97it/s]"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 5 -- Loss: 3.897, Improvement: 0.325, Motif Idx: 19, Pos Idx: 844, Time (s): 8.75\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"from tangermeme.design import greedy_substitution\n",
"\n",
"X_hat = greedy_substitution(model, X, motifs, y, output_mask=idxs, max_iter=5, verbose=True)"
]
},
{
"cell_type": "markdown",
"id": "7e524d59-5d7c-4822-973a-2147540becd4",
"metadata": {},
"source": [
"We can see what is happening during the process from the logs. `Iteration 0` is meant to show what the loss looks like before any substitutions are made. We specified that only three rounds should happen, but can also specify that the improvement must be above a certain threshold (`tol`, default is `1e-3`). The process will terminate early if adding in more motifs would cause a higher loss than the previous iteration. \n",
"\n",
"Unsurprisingly, by looking at the motif idx we can see that the AP-1 motif we know Beluga responds to has been chosen each time and that these insertions are somewhat close to each other. The middle is the sequence is not necessarily preferentially chosen by the design method, but it is possible that the models being used will prefer motifs to be in the middle of the sequence as an artifact of how they are trained. \n",
"\n",
"We can now take a look at what the predictions from the model look like before and after substituting in these motifs."
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "030dba20-5bf3-4faa-810f-e6ce5cefadac",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"from matplotlib import pyplot as plt\n",
"import seaborn; seaborn.set_style('whitegrid')\n",
"\n",
"from tangermeme.predict import predict\n",
"\n",
"y0 = predict(model, X).numpy(force=True)[0]\n",
"y1 = predict(model, X_hat).numpy(force=True)[0]\n",
"\n",
"plt.plot([-9, 5], [-9, 5], c='0.3')\n",
"plt.scatter(y0, y1, s=1)\n",
"plt.scatter(y0[idxs], y1[idxs], s=2, color='r', label='AP-1 Targets')\n",
"plt.legend()\n",
"plt.xlabel(\"Task Logits Before Design\")\n",
"plt.ylabel(\"Task Logits After Design\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"id": "774f60a5-0f09-4357-be16-d50c98ef1ae9",
"metadata": {},
"source": [
"Not only do we see that the tasks related to AP-1 binding have increased logits, we can see that they seem to be close to the desired value of 3. Interestingly, not all of the AP-1 tasks are increased to that level, and there seems to be quite a bit of a jitter in all of the other tasks. "
]
},
{
"cell_type": "markdown",
"id": "8623eb11-4db2-4256-a3b4-d001ced74a4f",
"metadata": {},
"source": [
"### Greedy Saturation Mutagenesis / Evolutionary Algorithms\n",
"\n",
"Rather than implanting entire motifs into sequences, some approaches consider changing only one nucleotide at a time. This is sometimes done probabilistically, where only a subset of potential mutations are considered, and is sometimes comprehensive, e.g., saturation mutagenesis. Regardless, each approach proceeds similarly to the greedy approach described above: a nucleotide is changed, the change in score is recorded, and the best individual mutation is made at the end of each round.\n",
"\n",
"This approach of only changing a single nucleotide in each round can be performed by simply passing in the four nucleotides as options rather than entire motifs."
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "e59f16bb-46c5-46c8-85c8-2bd285e0716e",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 0 -- Loss: 72.96, Improvement: N/A, Idx: N/A, Time (s): 0s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 2.92it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 1 -- Loss: 53.44, Improvement: 19.51, Motif Idx: 2, Pos Idx: 1058, Time (s): 1.374\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 3.00it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 2 -- Loss: 40.33, Improvement: 13.12, Motif Idx: 2, Pos Idx: 1083, Time (s): 1.337\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 3.00it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 3 -- Loss: 32.48, Improvement: 7.841, Motif Idx: 3, Pos Idx: 1087, Time (s): 1.337\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 3.00it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 4 -- Loss: 27.85, Improvement: 4.636, Motif Idx: 3, Pos Idx: 1029, Time (s): 1.335\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 3.04it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 5 -- Loss: 19.53, Improvement: 8.316, Motif Idx: 0, Pos Idx: 1031, Time (s): 1.317\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 2.99it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 6 -- Loss: 13.43, Improvement: 6.098, Motif Idx: 1, Pos Idx: 1028, Time (s): 1.34\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 3.03it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 7 -- Loss: 11.12, Improvement: 2.314, Motif Idx: 1, Pos Idx: 1035, Time (s): 1.324\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 2.86it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 8 -- Loss: 9.499, Improvement: 1.621, Motif Idx: 3, Pos Idx: 994, Time (s): 1.401\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 3.04it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 9 -- Loss: 8.413, Improvement: 1.086, Motif Idx: 0, Pos Idx: 1061, Time (s): 1.319\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 3.02it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 10 -- Loss: 7.222, Improvement: 1.191, Motif Idx: 3, Pos Idx: 1062, Time (s): 1.329\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 3.04it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 11 -- Loss: 6.564, Improvement: 0.6573, Motif Idx: 3, Pos Idx: 1086, Time (s): 1.322\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 3.06it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 12 -- Loss: 6.021, Improvement: 0.5431, Motif Idx: 3, Pos Idx: 1071, Time (s): 1.314\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 2.97it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 13 -- Loss: 5.359, Improvement: 0.6618, Motif Idx: 1, Pos Idx: 1070, Time (s): 1.353\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 3.04it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 14 -- Loss: 4.796, Improvement: 0.5637, Motif Idx: 2, Pos Idx: 1065, Time (s): 1.322\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 3.04it/s]"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 15 -- Loss: 4.527, Improvement: 0.2686, Motif Idx: 1, Pos Idx: 979, Time (s): 1.319\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"X_hat = greedy_substitution(model, X, ['A', 'C', 'G', 'T'], y, reverse_complement=False, output_mask=idxs, max_iter=15, verbose=True)"
]
},
{
"cell_type": "markdown",
"id": "0dda145f-2a1f-43fe-a1e7-c4ee6c41af34",
"metadata": {},
"source": [
"Each round is much faster because, although you are considering changes at roughly the same number of positions, you are considering a significantly smaller number of \"motifs\". However, you will likely need many more iterations because adding in a motif of length 8 will require 8 rounds rather than the single round in the other approach. Something interesting to note is that midway the improvement increases, indicating that an epistatic effect was discovered, i.e., that the inclusion of one or more earlier positions was required before that position could really improve output."
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "91ecc591-89f1-4c88-a2fd-c1a3b88380ce",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAksAAAGwCAYAAAC5ACFFAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAADFPklEQVR4nOydeVxU1fvHP8MMm7iAW4YLASauiJlpWsjXDYtvqdnigmW5BGV8W+yrrX5t+amlX8sWSNMWSVvMpSIlzS+SmUumiQuYQKDgggooMCwzc39/jOd6585dZ5/hvF8vXyN37j3nOefemfPM8zzneTQMwzCgUCgUCoVCoQji524BKBQKhUKhUDwZqixRKBQKhUKhSECVJQqFQqFQKBQJqLJEoVAoFAqFIgFVligUCoVCoVAkoMoShUKhUCgUigRUWaJQKBQKhUKRQOduAXwBk8kEg8EAPz8/aDQad4tDoVAoFApFAQzDwGQyQafTwc9P3H5ElSUHYDAYkJeX524xKBQKhUKh2EC/fv0QEBAg+j5VlhwA0Ub79esHrVbrZmkci9FoRF5enk+OTQl0/M17/ACdg+Y+foDOgS+Pn4xNyqoEUGXJIRDXm1ar9bkHieDLY1MCHX/zHj9A56C5jx+gc+DL45cLoaEB3hQKhUKhUCgSUGWJQqFQKBQKRQKqLFEoFAqFQqFIQGOWXIjRaERTU5O7xVCF0WgEANTX1/usr1oKZ47f39+/Wc4phUKheBtUWXIBDMPg3LlzqKqqcrcoqmEYBjqdDiUlJc0yh5Szxx8aGopOnTo1y7mlUCgUb4EqSy6AKEodO3ZEixYtvGphZBgGer0ewcHBXiW3o3DW+BmGQV1dHS5cuAAAuPHGGx3WNoVCoVAcC1WWnIzRaGQVpXbt2rlbHNWQ7KZBQUHNVlly1viDg4MBABcuXEDHjh2pS45CoVA8FBrg7WRIjFKLFi3cLAnFEyHPhbfFslEoFEpzgipLLqI5WmUo8tDngkKhUDwfqixRKBQKhUKhSECVJQqFQqFQKBQJqLJEoVAoFAqFIoFP7YYbMWIEysrKrI5PmTIFCxYssDq+b98+PPzww1bHf/zxR0RHRztFRm/jjz/+QHJyMoYOHYrVq1dbvHfmzBmMHDmS/bt169bo0aMH/vWvf+G2224TbTM9PR27du3CiRMn4O/vj99//130XH4fQsyZMwdPPfWUwhE5lpiYGHzwwQcYNWqUW/qnUCgUivPxKWVpw4YNbMZlAPjrr7/w6KOPYuzYsZLXbdu2DS1btmT/btu2rdNk9DY2btyIhx56CFu2bEF5eTnCw8Otzvn000/RvXt3XLp0CcuXL8fs2bPx/fffo2vXroJtNjU1YezYsYiLi8OGDRsk+7/xxhuxe/du9u81a9bgl19+wSeffMIeU7vTsLGxEQEBAaquoVAoFHvJ3FuC9JxCpCZEI3lIhLvFoajAp5QlvpKzcuVKdOvWTdLKAQDt2rVD69at7e6fq6hxjzEMw/7zJurq6rB161asXbsWVVVV2LhxI5588kn2fTKeNm3aoH379mjfvj0WLlyI4cOHY/fu3Zg0aZJgu8QKtHHjRot2hPDz80P79u3Zv4ODg6HVatljpaWlePXVV/Hnn39Cr9cjKioKzz77LIYOHcpeM2LECDzwwAMoKSnBjh07MHLkSCxZsgRff/01PvzwQ1RVVeGOO+7AwIED8eGHH+LAgQPstTt37sSKFStQVFSEjh07Yvz48UhJSYFOp8OIESMAgJ2T8PBw7Ny5E/n5+fi///s/HD16FBqNBhEREVi4cCH69etnNT7yXBiNRsHnx90QmTxRNlfR3OeguY8fcNwcpOecQllVPdJzTmHyoC6OEM0l+PIzoHRMPqUscWlsbMR3332HRx99VHZ79vjx49HY2Ijo6GikpqZiyJAhNvWZl5cneFyn00Gv18NkMtnULovBAE1xMZjISEDn/Fu3efNmRERE4KabbkJiYiLeeustTJ8+nZ3P+vp69rWurg7AdcWnrq6OPSZGY2Mjm8laKU1NTTCZTOw1ly9fxpAhQ/D4448jMDAQ33//PVJTU7Fx40Y2KzbDMFi9ejVmzpyJzMxMAMCePXvwn//8B2lpaRg+fDj27duH9PR0C3n27NmD+fPn4/nnn8eAAQNw5swZvPHGG2hqasLjjz+OtWvXYuTIkfjPf/6DoUOHQqvVoq6uDs899xxiYmKwdu1aaLVaFBQUwGg0Co6zoaEBTU1NyM/PVzwH7kDs2W5ONPc5aO7jB+yfg6SoAGzKb0RSVAAOHz7sGKFcSHN+BnxWWdqxYweuXr2KCRMmiJ7ToUMHvP766+jTpw8aGxuxZcsWTJ8+HWvXrsWgQYNU99mvXz+rLMz19fUoKSlBcHAwgoKCVLfJYjAACQnQ/P47mFtvBfbscbrC9P3332PcuHEAgJEjR2LhwoX4888/WasNGU9QUBBatGiBuro6pKenQ6vVYtiwYbLusYCAAGg0GlVuNH9/f/j5+bHXxMXFIS4ujn2/Z8+e2LVrF3777TckJycDMOcyGjJkCFJSUtjznn32Wdx5553ssV69euHYsWPIyclh2/70008xa9Ys3HPPPQgODkaPHj1QU1ODt99+G8888wx7Xvv27dGtWze27XPnzmHmzJno06cPK5MYfn5+8Pf3R/fu3e17PpyE0WhEXl6e4LPdXGjuc9Bcx79uXykycouQEh+Fh27t7JA5iIsD5jlORJfhy88AGZscPqssffvtt4iPj8cNN9wgek5UVBSioqLYvwcMGIBz585h9erVNilLWq3W6kHSarXQaDTsP5spLgauBUJrfv/d/HePHra3J0NRURHy8vLw3nvvATArKXfffTc2btyIYcOGmeW4Np7JkyfDz88Per0eHTp0wKJFi9CzZ0+8+uqr+P7779k2Dx06ZNEHuV7NvPCvqaurw/vvv4+cnBxcuHABRqMR9fX1OHv2rMW5ffv2teinuLgYo0aNsjgWGxuLnJwc9tixY8eQl5eHjIwM9pjRaERDQwPq6+vZciX8e/voo4/ilVdewXfffYehQ4di7NixFsoUfzwajUbw2fEkPF0+V9Dc56C5jT8jtxhlVfXIyC3GlMHmz6/cHPh6TFJzewa4+KSyVFZWhj179rALvRr69++P7777zglS2UlUFHDrrWaFadAg899OZMOGDTAYDBg+fDh7jGEY6HQ6VFdXo02bNuzx5cuXo3v37mjVqhXCwsLY4//6178wY8YMp8r51ltvYffu3Zg3bx66deuGoKAgpKWlWZUPIYoNdyx8JY0fO2UymfDUU0/hjjvusCqkGxgYKCrTU089hX/+85/YtWsXcnNzsWLFCixfvhyjR4+2dZgUCsXFpCZEs4qPUtJzClFWpUd6TqFPKkvNGZ9UljZu3Ih27dohISFB9bUnTpxAhw4dHC+Uveh0wG+/AUVFZkXJiS44g8GALVu2YP78+Rg6dCjq6+vZQrJPPfUUvv/+e9bFBZh3rAlZTtq1a+f04sEHDx7EhAkTWEWktrZWMH0En6ioKCvT69GjRy3+7t27N4qLizFlyhS0aNFC0ALm7+8vGCAYGRmJyMhITJ8+Hc8++yy+/fZbqixRKF5E8pAIVuFRGgRsi4JF8Q58TlkymUzYuHEjxo8fDx1PoVi2bBnOnz+Pt956C4A5JqVLly7o3r07mpqa8N133yE7O9smi5RL0Omc6noj5OTkoLq6Gvfffz9atmyJuro6VlkYO3YsNmzYYKEsqaW8vBzV1dUoLy+H0WjEiRMnAADdunVDSEiIqra6deuG7du3Y8SIEdBoNHjnnXcUBdInJycjOTkZn3zyCf7xj39g7969yM3NtVCInnzySaSkpKBdu3a455572GDtgoICPPPMMwCAzp0747fffsMtt9yCgIAABAYG4q233kJiYiK6dOmCc+fOIS8vD2PGjFE1LgqF4n1wFSyKb+FzytKePXtQXl6OiRMnWr1XUVGBs2fPsn83NTVhyZIlOH/+PIKCgtC9e3esXLnSwvXUHNmwYQOGDh2KVq1aWbmmxowZg4yMDBw7dszCFaeGFStWYNOmTezf48ePBwB8/vnnGDx4sKq2XnjhBbz44ouYNGkSwsLCMGvWLNTW1speN3DgQCxcuBDvv/8+3nnnHdxxxx2YPn06vvjiC/acO++8E+np6Xj//ffx+eefQ6fTISoqCg888AB7zrx587B48WJ88803uOGGG7Bt2zZUVVVh3rx5uHjxIsLCwjBmzBikpaWpGheFQnE9vh5zRLEdDeNtyX88EKPRiMOHDyMuLk5wN1xxcTEiIyM9creTHGQrvZgbypd4+eWXUVRUhHXr1rHHnD1+T38+pJ7t5kJzn4PmNP5hi3eirEqPzqHB+HX+CPZ4c5oDIXx5/ErHRmvDUZotq1evRn5+PkpKSrB27Vps3rxZMtUEhULxXTL3lqC2wYDQYH8ac0SxwufccBSKUo4cOYKPP/4YtbW16Nq1K1566SULFxuFQmk+pOcUokrfhM6hwR7pgqMuQvdClSVKs+Xdd991twgUCsVD8PSdbDQtgXuhbjgXQUPDKELQ54JC8QySh0Tg1/kjVCkimXtLMGzxTmTuLXGiZGZSE6LROTTYY5U5X4cqS07G398fAFTVP6M0H8hzQZ4TCsXdOFIBcKUyoaRfR8vDtfY4G1uUOYrjoG44J6PVahEaGooLFy4AgNftKmMYBg0NDfDz8/MquR2Fs8ZPdtlduHABoaGhPrfDhOK9ONLd4y7XEel3wRZzolnSt6PlUeu6o3FH3gtVllxAp06dAIBVmLwJhmHQ1NQEf3//ZqssOXP8oaGh7PNBoXgCjozdcVccUGpCNBZsOQojAwvFSEweW5UYtUkoadyR90KVJReg0Whw4403omPHjlY1yzwdo9GI/Px8dO/evVlaP5w5fn9//2Y5pxTPxpFZqN2V0ZprSeIqRtzj/L/tUWKUKlueHkROEYcqSy7EGys2k5pIQUFBXie7I2ju46dQnIkz3VJiipqQYsRXYtTKpVTZouVQvBca4E2hUCgUt+DKAGkS3D0wIsxqVxk/eFqtXHSnmu9DlSUKhUKhuAU1Soa9O9mWZhegrEqP3JMVsrvK1Co/UjvV3LUjkOJYqBuOQqFQKG5BjVtKqauL60Ij16m1+DjSXUaDun0DalmiUCgUiijr9pUiJesC1u0rdascSq09XOWE+/+5iTHoHBqMuYkxivskViGpsctZjqiLzjegyhKFQqFQRMnILUJFnQkZuUVulUPI1SWkqHCVE+7/bUnqSJQtqbHLxTfRZJK+AVWWKBQKhSJKSnwUOrTwQ0p8lLtFsUJIUeEqJ/YqKkTZ4o+dq6RRy1HzgMYsUSgUCkWUKYO7oXfgZcTFdXNou3Lb8zP3lmBpdgEAYG5ijOA5zs5bRBQuo9GIw4cvs8e5Shq1GjUPqGWJQqFQKC5Hzn2VnlOIKn0TqvRNilxcaurB2btDzZW7+CieAbUsUSgUCsXlyFmFUhOiWctSakK0oKWJa53i7zoj79U2GCwULv4xJVahdftKsWLHBaQ1lGLa0Ein7OKjeDbUskShUCgUlyMXT5Q8JAKHF4zB4QVjkDwkQtDSxFVE+NYe8h4A9rjQMSXYE+TOlYtambwXqixRKBQKxWYcqQBItZWaEI3QYH+EBvuzSo7Ubjfy3tzEGPb4wIgwaDVAfI8OqmKN7Aly58rlyozlFMdC3XAUCoVCsRlHupmk2hJyfUm5w4TeO1hSCSNjfnUHtJCu90ItSxQKhUKx2ULkyK3ztrYlJ7tUXTglKHHDKZk/mnPJe6GWJQqFQqGoshDxt/07avG3tS052ZdmF6BK34TaBgMOLxijuv2U+Cis2JEv6Yajgdy+DbUsUSgUCkWVVcdVsTdKrV3OTgw5ZXA3ZCR1xJTB4rmmaHJK34ZaligUCoUiaNUxb9fPh8FoxLxr2+YB18XeKLXWyFmk5ibGOF1eW61icsk5KZ4BVZYoFAqFIoh5u74BgDluhyhLjnS9iZG5twS1DQaL3W+24gp5xUhbfwhZR8qRFBuOFZMHWL1P3XfeAXXDUSgUCkUQ83Z9HVoGaCTjdfjuMkekEyB5lUICdaLlULwhZ1HWkXIYGfOrENR95x1QZYlCoVAogiQPicDBl0fhs3E3WMXrcJUVfgyTI2Ka5JSIpdkFKKvSY2l2gWLFyRlKnVwbSbHh0GrMr0LQHXLeAVWWKBQKpZkjteCv21eKlKwLWLev1OK4VPZspdYS0m/a+kNW/atRIpQqZ9zzMveWYMGWo3YrdXJ9r5g8AIWLkgRdcErxFiuaL0OVJQqFQvER0tYfQvQLWUhbf0jVdVILPjfHEHfRlsqerVTRIf1mHSlXrbTMTYxhM3QrLSnCPS89pxBGBtBqYJcLzBVuNJr52/1QZYlCoVB8BLn4GDGkFnxuqQ9+MDJfIRJSVJQoL0mx4ZIKh1Ab3P6VlhThnkf6Xjiur10uMFe40Whck/uhu+EoFArFR0iKDWd3XvGR2qJOFA6ilHDPmTK4G3oHXkZcXDdo/Pwkt+AL7ewSimfiJrNU62aTO19pWgO1O+SyC+uQtj0HqQndXR5f5M7dfBQz1LJEoVAoXgrf4iIVH6PElSN3jpwVRcgCwnd92eJOUmNZcZalZ1N+Dcqq6qkrrJlCLUsUCoXipTjK4kKsTgMjwthzHQXfKsKVQWlCRk+wrEzo2RJZRY3UFdZMocoShUKheClqMmlLKRxE6QKAX+ePsFkevvImVEOOrzjZm5DRVRmwE6NbYN7EodBqtU7rg+K5UDcchUKheCmOcjmpDSAWC+TmZ9yWc7s5InCZ7hSjuAKfUpbee+89xMTEWPwbNmyY5DX79+/Hfffdh379+mHkyJFYv369i6SlUCjNHU/Jn6NW6RJSUIQybsspQ45Q9qT68JT5pXg/PqUsAcDNN9+M3bt3s/++//570XNPnz6N2bNnY+DAgdi8eTNSUlLw5ptvIjs724USUyiU5ootVhFXKgCZe0sQ/3YOsgvrLI7LBXITlChD9o5Hqg9nWJ2oAtY88bmYJa1Wiw4dOig698svv8SNN96Il156CQAQHR2NvLw8rFmzBomJiar7NhqNqq/xdMiYfHFsSqDjb97jB5w7BynxkcjILUJKfKTi9tNzTl3blXUKkwd1sXp/3b7Sa21GWZUoUQvpa1N+I+Zy5Js8qAvbN5Fb6JiaPsTGYw+2zK8Q3GfAmfJ6Ku78HjCZTPjf//6H/Px8JCcno02bNg5tX+mYfE5ZKikpwR133IGAgAD0798fzz77LLp27Sp47uHDh63cdHfeeSe+/fZbNDU1wd/fX1XfeXl5Nsvt6fjy2JRAx9+8xw84Zw5On6lDY2MjTp85g8OBlxVdkxQVgE35jUiKCsDhw4et3l+x4wIq6kxYsSMfvRW2KdfXhJ4tnfYMyI3HHnoHAitGhwK4jMOH7ZsLwPwMOFNeT8fV3wOnT5/G1q1bcfbsWQDADTfcgO7du7tUBoJPKUuxsbFYsmQJbrrpJly6dAnp6emYNGkSfvjhB4SFhVmdf/HiRbRv397iWLt27WAwGFBZWYmOHTuq6r9fv34+t1PCaDQiLy/PJ8emBDr+5j1+wLlzkLY9BxV1JmQVNWLexKGKromLA+ZJtdlw3bIUF2efZSkuDphrx/iVWLnkxuMJcJ+BuDitx8vraFz9PXDp0iV8/vnnyM3NBQC0aNECkyZNwj//+U9oNBqH9kXGJodPKUvDhw+3+DsuLg6jR4/G5s2b8eijjwpew594hmEEjytBq9X67ILiy2NTAh1/8x4/YDkHjtqunprQnW3HUfM7bWgkpg2NdEhbXGx5BjJyi1FWVY+M3GKnyORqlM6Bq9IZuBpnfw80NjZiy5Yt+Oabb1BfXw+NRoNRo0Zh2rRpCA0NdVq/SvC5AG8uLVq0QI8ePfD3338Lvt++fXtUVFRYHLt8+TJ0Op3bbwyFQvFcHBU47Iq6YoD7gpK9paaZmvkRO5d7nKYzUAfDMNi3bx/mzJmDtWvXor6+Hj179sSyZcvw1FNPecR67NPKUmNjIwoLC0UDvuPi4rBnzx6LY7t370bfvn1VxytRKJTmg7coAQR3Ld6uUgbtRc38iJ3LPe5tz4c7OX36NP7zn//gzTffxLlz59C2bVs8++yzWLJkidvik4TwKWVpyZIl2L9/P06fPo0///wTaWlpqKmpwYQJEwAAy5Ytw7///W/2/EmTJqG8vByLFi1CYWEhNmzYgG+//RaPPfaYu4ZAoVC8ACElwJO3lNu6eIulDvA11MyP2Lnc496iJLqTmpoafPzxx0hLS8OhQ4eg0+lw//33Iz09HQkJCQ6PTbIXn4pZOnfuHJ599llUVVUhLCwMcXFx+Prrr9G5c2cAQEVFBRtVDwBdu3bFypUrsWjRInzxxRfo2LEjXnrpJZvSBlAolOaNI0p3OAtba6uZx2ROHeDtQc38OCKhUixKEDvXE+rXeQNGoxE///wz1q5di+rqagDAbbfdhhkzZuDGG290s3Ti+JSytHz5csn3Fy9ebHXstttuw6ZNm5wlEoVCaSaoqdPmLZjHdApJUQFuk8FRwdJLswtQpW/C0uwCJA+JkFRufTVA290cP34cq1atQmGh2YXZpUsXzJo1CwMGDHCzZPL4lBuOQqFQ3IUjXS+e4tJLHhKBlPgobMqvwbp9pZLnOktmZ8VbSbneaIC2Y7l06RKWLVuG+fPno7CwEC1atMCMGTOwYsUKr1CUAKosUSiUZoinKCNieNJinZFbhIo6EzJyiyTPc4dSo4a5iTHoHBqMuYkxAKSV24ERYdBqzK8U22lsbMTXX3+N1NRU7Nq1CxqNBmPGjEFGRgbGjRsHnc57nFveIymFQqE4CE+OLwI8y6WXEh+FFTvykRIfJXmes2R2VCyQmnYOllTCyJhfKeohqQDWrFmDc+fOAQB69uyJ2bNne9QONzVQZYlCoTQ7PEkZEcKTgoWnDO6G3oGXZbOBE3mJZclR8qetP4SsI+VIig3HisniLhtunBGRw9aYI098Prwljur06dNYtWoVWwqmbdu2mD59OoYPH+5xO9zUQJUlCoXS7PAkZcSXcIbFLutIOYyM+VVKWeK7Ae2RwxOfD0+3htbU1ODLL79EVlYWjEYjdDodJkyYgPvvvx/BwcHuFs9uqLJEoVAoFIfgDItMn/A2OFJWjT7h0tXm+X17mmXIXjzR2gWYUwHs2LEDmZmZbCqAwYMH47HHHvPoVABqocoShUKhUByClEXGVjfSpdpGi1elfXuK9UXtuMXO90RrFz8VQNeuXTFz5kyv2eGmBrobjkKhUCiSZBfWIf7tHLt2D9q6W86TSoes21equobc0uwCVeN21K5CZ+745KcCCAkJwYwZM/Duu+/6pKIEUGWJQqFQKDJsyq9BWVW9XQu4rUpP8pAI1gXl7lQPGblFqmvIAVA1bkcph85I5SCWCiA9Pd3rUgGoxXdHRqFQvA+DASgqAqKiAO4Xr8Bxb9kd5C1IzeeEni2RVdRo1wJujxvJU4KbU+KjkJFbrLiGnC3Pp6PcbY6McWIYBvn5+cjIyMD58+cBeH8qALVQZYlCoXgGBgMuxg5E+xNHcLFXLNofOWhWjESOe8oC6iuQ+Vyw5SgAy5ifxOgWmDdxKLRarVtkk1r4HRUTpIQpg7th2tBIRedylR5b+rT3x4CjlK7S0lKsWrUKf/75JwBzKoBHH30U8fHxXp0KQC3UDUehUDyDoiK0P3EEAMyvRUWSx50Vy+Kq7N5i/bgru3hqQjS0GsDIwGGuG0eNRSrbtlp3k5Lzidxp6w9h2OKdsqVeHNGnkmtc+WzU1NTg448/RlpaGv78809otVpMnDgR6enpXp8zyRaoskShUDyDqChc7BULAObXqCjJ446sxcbFVaVGxPpxdakTsgADwMJxfUUVUDXBzQSxscgt+nxlRapPrtKsRJlQomQTubOOlKOsSi9b6kUOsT6l5BW6xhXPhtFoRHZ2NlJTU/Hdd9/BZDJh8ODBePLJJ5GcnOwTOZNsgbrhKBQ5DAbg5Eng5pst42gojkWnM7vYiorQnhuzJHbcSbgqn41YP67Op8NdgKWUT3Nwc70qt+fAiDCcq9Zb1ViTc6GS989V62FkgDezTmDBlqOCWby57qZhi3fKumaVuKfIPRgYEYaDJZVIiY8EcFnRmNX0KTUPQtc4+9kQSwUQGxvLZuRurtBvfgpFCoMBN06ZDm1RvmUcDcU56HRAjx7KjzsBV+WzEevH1fl0lC7AaoKbCWI11uT6JO+3CwnAsfJq6JuMAKSzeGfuLUFtgwGhwf52KxP8e2A0GnH4sO3KkhhqlR9nPRuXLl3Cp59+il27dgEAQkJCMHnyZNx9993Q6XQwGo0O79PboN/6FIoURUUIL8oHwImXcdGiTaG4AqULsJrgZoKYMiDXJ3l/2OKdMDJAsL8WjQYjkmLDrc4lgdC1DQZU6ZvQOTRYdjyespOSjJO441wtT2NjIzZv3owNGzagvr4eGo0Go0ePRnJyMkJDQ10mhzdAY5YoFCmiolAe1RMAL46GQhHAXcHZavt3lpz8dvlxZWr7JXE7LyX1wsJxfXGwpNIqjklJPiN+v2pifzL3liD+7RxkF9YpktkWXB2nxjAM9u7dizlz5iAzMxP19fXo2bMnli1bhjlz5lBFSQCqLFEoUuh0OLvuUxiPH6cuOIosrl70bO3fXjnFlB7S7tLsAsn3hfoVCurmKlv8oGvSBlGo5ibGiCav5PerZiel+dp6bMqvUTVH/HHZG3TuKEpLS7FgwQL83//9H86dO4e2bdviueeew5IlS5pNziRboMoShSIHiZehihJFBneX5lDSvyNie8SUHtJ/g8HIKk1K5RNShriKBrk2KTbcog0hhUpMLqFr5BgYEQatBohpF6BqjvjjklJMnbWzk0tNTQ1WrVqFtLQ0HD58GDqdDg888ECzTQWgFvrtT6FIYTAgsKQE6NsXUJqQTywLNcXnsTcAVyqWRkmcjZL+03MKJWN7lPQjF4sUt/An6JtMquQjO+f6hLfBpdpGtg8lO/WUymULJEi94JJ0IV+1MrkKo9GIHTt2YO3atbhy5QoAYMiQIXjsscfQqVMnt8jkjVDLEoUihsGAyrhB6DtxIirjBpmVIAXXXIwdCMTEmF+VXEOhXEPKCuEoF5+c9cmefoglKL5HB9Y1JncucU8RpeRSbSOrGKm11HEtNI5K+mmWIQgTerZUdL6UTK7m+PHjeO655/DBBx/gypUr6Nq1KxYuXIgXX3yRKkoqocoSpflC8ieJKTRFRWifnwcA5tciBYnpxLJQUygKkFIObHXxyQVdC/UTGuyP2gaDqEIhl1DzYEmlVR9yQdbE3cXNyWSPouGopJ/JQyKQ+3wCEqNbqJYBcE/Q/8WLF7Fs2TLMnz8fRUVFCAkJwcyZM/Huu+9iwADh1AsUaaiyRGmeKLEARUXhYs9+AICLPRXuhBPLQk2hKEBKOVCqONiz84v0ExKoQ5W+SfQaMcVNSUySWJA1sSzlnqxwiHJhi4zOwJVB/42Njfj666+RmpqKXbt2QaPRYMyYMcjIyMC9994LHQ0LsBmqLFGaJ0osQDodwg4fwNFvv0XY4f3K4o9ItumCAvndc3KWLQoF6i0T9uz8IshdI5YSAICoQscvS8KPiyLvA3CIciGmXLraLeYK5YykAnjyySeRmZmJhoYG9OrVC//9738xZ84ctGnTxml9NxeomklpnlyzALGV7MUsQDodGiIi1AVqK8k2fc2yxfZP0xJQRJArDcKHH1AsFNzMV1a4f5M+yXtKsjcrkVGuLAk3QaNUQLSU7O5MMCmGszOyl5aWYtWqVfjzzz8BAO3atcP06dMRHx9Pd7g5EGpZojRP1FiAnIG3xzZRq5jLsCfIWQy+9Yn8vWDLUSzNLlBt2ZGTkW8ds8faIia7u3JbuQtuKoA///wT/v7+ePDBB/Hhhx/SVABOgCpLlOaLO/MneXNsE93x53S4yoUz3EZ8ZSU1IRpaDWBkzO+rVWTkZOQrNFLnyyk/QrK7M7eVsxBzvxqNRmzbtg0pKSn4/vvvYTKZMGTIEHzwwQdITk5GcHCwmyT2bajdn+K9eHM+I2LZKioyuwBtld8dcyBkFXNDvTxPd7/Yg1rXmxL488Vtl/zfWfOpJteQpxSXdTdCz8Dx48excuVKFF2zRHft2hWzZs1CXFycGyVtHlDLEsU78QXrhr2WLXfNgbOtYgpdfN7gfrF127gzrCVy8+XMwGc1bZP8SkJlSwDPue/OTgnAfQYuXryIpUuXCqYCoIqSa6DKEsU78faYH0fgrjlwZryXCgXQU9wvUoumrQu7vYpL5t4S9HplKyLnZyFt/SEAzp0vRysOUvPG31XnrsLFzlbakodE4H/P3oGA0n1ITU1Fbm4uNBoNEhMTaSoAN0CVJYp34s0xP47CnXPgrHgvFQqgOzMjc1G6sNuLGsUgPacQ+iYTGABZR8oBOHe+xObAGZY1JbXgXIEzlU+GYfDbb78JpgJ48sknaSoAN0DVUop34qiYH2/GnXNQXw/s2gUMHw4EBTmuXaUpHTwIqRgbR8bTqIljSk2IxptZx1HfZEJSbLhD+pfrT2gObI29UjpvSuObnBHf5qxYKZoKwDNR/e168eJFLFmyBL/99hsuX74MhmEs3j9x4oTDhKNQJFGSz8jXcccc1NejoU0YAhvr0RAQhMDqSscpTF6oBDti0VRbvDZzbwmWZhcAAOYmxggmXuTmICK76pT2xSe7sA5p23OQmtBdNLu4WCJKpcHatsildO6dETDvaGpqarB+/XpkZWXBZDLB398fEyZMwP33348gR/4godiE6m+i+fPn4+zZs3jiiSfQsWNHZ8hEoVA8mV27ENhYDwDm1127gMRERZcqWhCbkRJM5qO2wcCWF1Ga1LFK3wQAktcI5SSS6kvs/mzKr0FFncniGiX3Uo0i6UyFRu0OO1diNBqxfft2ZGZm4sqVKwCAIUOG4LHHHqPFbj0I1crSwYMHsW7dOvTq1csZ8lAozQNvTnswfLjZokQsS8OHK77UG37huxIyH6HB/qriX1ITolnLktA1RJEhRWmJskD6Cvb3Q3mVHmnrD2HF5OuFVck5S7MLWOVi8qAumNCzJbKKGi36ElLEpBQnOeXKmQqNp6YXoKkAvAfV39I33nijleuNQvEq3K2oeHupk6Brrrddu8yKkgoXgSf/wncWXMXlYEmlVT00KQVCTMGQW/yJIgOYa7Vxj6cmRGPBlqNs8DdXWSLy1DYYWEVo8qAuSIxugXkTh0Kr1Vqdy1XE+IoT9/9CirJU7id78eQ8XBcvXsSnn36K3NxcAEBISAimTJmCu+66i+5w81BU74Z78cUXsWzZMpw5c8YZ8tjFRx99hIkTJ2LAgAG4/fbb8cQTT7Aauxj79u1DTEyM1b/CQs/N3UKxA0/Iz+TtaQ8MBqC0FBg5UnWskq07sty5RVwpRMa09YdYWTP3lmDBlqMoq9Lj+z/LWasNgcwHAMHx2brbS2inFnfuk2LDodXAKvibnDM3MUbW0sVtj9sfKZdCrFNEfiGZuJYs7vgdcb89JR8Tl8bGRnz11VeCqQDuueceqih5MKrvzDPPPAO9Xo/Ro0cjKCgI/v7+Fu/v37/fYcKpZf/+/Zg6dSr69esHo9GI5cuXY8aMGcjKykKLFi0kr922bRtatmzJ/t22bVtni0txB56QfdoLd3yxuMkq5g3uOyLjuWo9jAzYRdrIAFoNEKDzg77JxJ7PtXyIjU+JJU4o2FvKSpO5twS5JyvQKsgft0UKf89xr1+7pxgrdlxAWkMppg2NlD2fqwzyLWd8mYQsWfyUANxr1FiLPMmKyTAM9u7dizVr1uD8+fMAgF69emH27NmIjna/fBR5VH/Lvfjii86QwyGsXr3a4u9Fixbh9ttvx7FjxzBo0CDJa9u1a4fWrVvb1b+S6tzeBhmTz4wtIgKVPfuhfX4eLkfFoE3nzoDE2Jwyfo0GYYf2w1hUhLCoKBg1GkkZ3InV+P/6y0LZNP71l0uUzZT4SGTkFiElPtLlz2Lm3r/x/s8XMKf+byQPuUn0vJT4SCzd/hcaDUYE6LRIiY/E/r8v41y1Hnf364Tbbmp7bQxRMBqNSM85hbKqeqTnnEJKfJTg+CYP6oLJg7oAMCstS7f/BQ0YPDe6B6YM7gYAWJqdjyq92UKannOKPV+M9JxTnOBwBefvKjIHd+8qYvuU4rnRN7Pj5MtPjpN2yPvr9pVajF/sfnPnTE5ubt/2PjP2fA+cPn0aq1evtkgF8Mgjj+COO+6ARqPxiu9Wn1sHOCgdk4bx4QCkkpISjBkzBt9//z16iHyh79u3Dw8//DA6d+6MxsZGREdHIzU1FUOGDFHcj9FoxOHDhx0kNcWhGAwILCtDQ+fO1y0g9fXoOvlhdDxdhPKonji77lPvihlyJwYDbpwyHeFF+SiP7omzX3zqk3OXXViHTfk1mNCzJbsTrEMLP2QkdbR6H4DouSlZF6yuFeojMVra8g2AbQuARXuPbDmPmkYGGgCzbmnNtpVdWId1eVcBDTClbyvZ40rmQomccvJ3aOHHzpXaNpfvrcKeM/UY2iUIzwwJtVkWV6DX65GTk4MDBw6AYRhotVoMHToUd9xxBwICAtwtHoVHXFycRUweH9XKUk1Njeh7AQEBHvMQMAyD1NRUXLlyBevWrRM9r6ioCL///jv69OmDxsZGbNmyBV9++SXWrl0ra40iEGWpX79+kpPtjRiNRuTl5Xnn2AwGVMYNQvv8PFzs2Q9hhw+YF/aTJ6Ht3Zs9zXj8uKh1xKvH7wAEx+/uAHkXEP92Dsqq6qHVAHf1vQH7CiswZ2QP1rLEfb9VkA5VegO0GqB3eGscL7/CWpKWbT8JBhrMHX2zIqvM018dxo9553B3v05456E4i/fW7SsVtCxdt8pEWfRBZAQADYA2wTqL65TiqM8AV86M3CKUVdWjc2gQcp9PUNwGGZPa6+xFzRwYjUb8/PPP+OKLL9hUAIMHD8b06dO9NhWAL38PkrHJKUuqv+luvfVWySyinTp1woQJEzBnzhz4+bmvmsprr72GkydPSipKABAVFYUoTszIgAEDcO7cOaxevVqxskTQarU+9yARvHJshYVon58HAObXwkKgd2/g5pstY4ZuvhmQGZtXjt+BWIxfqwU8MHWII3c/pSZ0x4ItR2FkgEOnq5GR1BFxcTexc8B9H9BAqzHHJh0vvwIjA/xRWo0/SqtRpTegc2iwVawPf4cceS2v0oMB8GPeObw3xfJ5mzY00qId7nh/nT9ScAxLswtQrW8CA6BKb0BGbrFo3JEc9n4GuPJr/PxY2eXa5I4zNaG74uucgdwc+HoqgOb8PahaWVq8eDGWL1+OCRMmIDY2FgzDIC8vD5s3b0ZqaiouX76MNWvWICAgACkpKc6QWZbXX38dO3fuRGZmpk2afP/+/fHdd985QTKKS4mKwqWe/dDumsJ0ceIktM/7wzVZopuB9cXTcGQQOLk+PacQKfGRAC5fs4wUWyhjROHJPVkBAIjv0YFND7C/+DLKq/S4XNvI7uoiC/3S7AJU6ZtY5Yhs8w/290OjwbpEiVD6ATLeBVuOWsjMPX9uYgwASOZk8nS499UTagEKQVMB+D6q7+KmTZswb9483H333eyxkSNHokePHvjqq6/w2Wef4cYbb0RGRobLlSWGYfD6669j+/btWLt2Lbp27WpTOydOnECHDh0cLB3F5eh0aPftl0CfPgCuWZfI7jdnZon29jxKXorSnWNKrU9k95bRaMSSb89g9aFz7C437o4zkk1bqwFui2zL5i1KzykEA0DfZGR3xpFt8tXXAqyDrilHZMfc6N6dcLCkEgDQ65WtqG8y4Z7+4ThYUmm1047kS+LKxM8IvmDLUSwc1xeHF4xxwAzbDn/e1da5U1rmxVZ5bKWxsRGbNm3Chg0b0NDQAI1GgzFjxiA5OZkWu/UxVPvJDh8+jN6ceA9C79692SDngQMH4uzZs3YLp5aFCxfiu+++w7JlyxASEoKKigpUVFSgvr6ePWfZsmX497//zf796aefYseOHfj777/x119/YdmyZcjOzkZycrLL5W92GAzAyZPOzXXUowcu9ooFAPOrK7bpe3seJS9FLmdR5t4SvLL5qFWuIzFIrp91+0qxKb+GVWj4ylhqQjTrhuPm9ElNiEawvx80AAZGhLF5hgCAgbmtl5J6Y+G4vugcGoyF4/qySlHWkXLom0xs4khybVJsOJurKHlIBHstNwEksVIRmRZsOer2/FT8nEdCOZfE4OZzSs8pRJW+iS3X4ih51MIwDH777Tc88cQT+OKLL9DQ0IDevXvjv//9L5588kmqKPkgqn/udurUCRs2bMDcuXMtjm/YsIF1eVVVVdm9Dd8W1q9fDwCYNm2axfFFixbhvvvuAwBUVFRYKHJNTU1YsmQJzp8/j6CgIHTv3h0rV67EcBUlHCg2oNT6Yq87yx2FWb05j5IPIGa1IJYePmJWBtJORm4Rp9xHdyt319LsAvj5aWAyMmgXEoBhi3ey7rJAnRb6JhO2Hz+P7/8sR5C/H0b37oTckxVoMBixNLsAcxNjLLJsL80uQIPBCIORQZOJQYDO/JuWew6BWzA3buFPaDAYERrsz7rf+JYnZ8GfQ/7ffKufrdm65cq8qGnH1hxMpaWlWLVqlUUqgEcffRR33nmnZDwvxbtRvRvu559/xr/+9S9ERUWhX79+0Gg0yMvLQ1FREVasWIF//OMfWLduHUpKSvDCCy84S26PguyGk4um90acNraTJ4GYmOt/FxRYu8Uc7c6yQfGyefw+ErPkjc+2mPIj5sIZtngnyqr06BwabKGQkHZS4iPRO/CyxRyQ9y7XNlgkmtTgutXIyAChweakvSSvEQDWukQsQPx+4xb+hCp9E0KD/RESqBOUjT9OrkWJe6697ibz9aeQFBVgVe6Eew5RykjfYnPqrRiNRvz22284duwYtm7dCpPJBH9/f0yYMAH3338/glRmsvc2vPF7QClKx6baDTdy5Ehs27YN8fHxqK6uRmVlJeLj47F161b84x//AABMmTKl2ShKFBu5Zn0BJNxjjnRnubrMCYmJ8mJFKXNvCeLfzkF2YZ3lG65wnwrIorT8hVhJleQhEZibGIOQQMt7wnUJcfsh7QhttSfKST1HUQLAWq46tQlC59Bgq/6C/f2u7eqKRmiwP0KD/SWtG0S2gRFhVqVU+KVElLSnFnMf9diUb5kyhjtP6TmFVi5KNW42T4P/rBmNRmRnZ+O9995DVlYWTCYThgwZgg8++ADJyck+ryhRzNj0Td6lSxcrNxyFogol7jFHurM8ocyJl3F9oWzEPHLQi8qdCFlVhNrhuoSIRYTfT3ZhHdK257BuuIERYThXrUef8Da4VNuIgRFh+O7Pcvb8c9X1KFyUZCE/v7AsCbomizORc25ijFWJEGJtIrvnuBYlfikRbntq5407ZwBQ22BAm2B/TOgZLHo/5IoBexvcsQ1oVYOVK1eiuLgYgDkVwOzZs9G/f383S0lxNYosS/n5+TCZTOz/pf5RKIqRs74QhaqgwP5FWYkly5m4wRqjFv4varN1IIjNUg3AbcHrtlgqhIJ4xdohMT+XaxsErTOb8muuldkwt3WwpBJGBrhU24hf54/AiskDcG//69v9k2LD2fkErscbkaK6XJn4ckoVG9ZpzTmdBkaEiZ4nVrxWzjrHLfqbnlPIpjfQgLHKss1tV0gOVxexdWSh5dSEaHQNNiCuKhcvvPACiouLERISgrFjx2L58uVUUWqmKFp9xo8fj19//RXt2rXD+PHjodFoIBTqpNFocOLECYcLSWnGOGqLvzsCvQlekkqAb4VIHhKByYO6WJbycVPwui0BwUJBvGLtEMUAANqGBFqdE9MuAJf19RgYESba9orJA3BbZFuk5xSyr2TnHdnOz3dXZe4tweXaRgDAhSv1iFv4k+CWeGJtIikBSHoBIYjViyhUACyuFbMy8d1pJL7LXEjFErn74eoito7KsdXQ0AD/kr3odHQDTl9LBZCYmIhJkyahuLjY5+J1KMpR9I39888/o23btuz/KRSvxJm5laTwEhegogVOQul0ZAZtR2DrjqvaBgPS1h9ikz8yJhP2nKmHiQGrpMgpXWSXG1FSyqr0CA32t7DGAOZFXt9kLuTZZGJElRnurje5e0SsXllHyi2UNm7/Qgi507hJOdXAnx9nPxtyz65c/yQVwJo1a3DhwgUA5nQ4s2bNQnR0tKoCsp72OaA4BkXKUufOnQX/T6FQFOAlqQQUKxciSqcjM2irQWpxUrpwxffogKwj5QjQaVGlb0LWkXJO3iQGpmuGdGJZEpOjmrPrTUjB4cswMCIM5VV66PzM1huDiZHsQ+oecbN885NXKp0f7vvcpJyHD6tTlvg4+9mQe3al+v/gu9/wwzeZ8K8+DQBo3749pk+fbnMqAHd9DijORfVuuE2bNiEnJ4f9+6233sKtt96KSZMmoayszJGyUSi+gSNjrzwYW3dA2RtvIhUfozR2hlhjAnV+CA32R8C119SEaKTER7GOKFLWREj+N7OOg4E5fcDcxBirmCV+AHbm3hIcLKkEA6Bj6yB0bB0E5posUnNC4qviFv5k8T4Z68GSSotklVIxUK6KLXL37jih/mtqzMHb2z5eDP/q02A0Wjz00EP48MMPER8fb3POJHePleIcVCtLGRkZCAwMBAAcOnQIX3zxBZ5//nmEhoZi0aJFDheQQvEJfCCVgBxSi7IUcgu2nDIltTgpXbjIefE9OuBqfRP0TSaEBOqQPCQCUwZ3Q4i/+MK5NLsAZVV6Nt9Sm2B/q5IeZBxxC39is4hzt/xfuFqPsio9gv21VrvYuNfGLfyJdfXxs1jbskjzrxGa63X7SpGSdQHr9pUqbpffllx2dWfDfTaNRiO2bduGxx9/HD/88AM0YNDQtjuSUl/G1KlT7UoFQF1wvotqZencuXOIiDA/BDt27EBiYiIeeughPPfcc/j9998dLiCF4hF4wW42b0VukZdTprgZovmLsNJFmpyXe7ICRsZsHeLKM6VfK3QODWIzYwsR7G+2RtU2GhA5PwvtQgKsSpFU6ZtY61NtgwH7iy/jan0TmoxmP1+jwciOh5tfiasgARDMqcQvC6LEYsSfO6HrMnKLUFFnQkau9M5HvqIl1Jard8nxOXbsGJ599ll8+OGHuHr1Krp164bXX38d2Z/+FyljB9rdvrvHR3EeqpWlFi1aoKqqCgDw66+/YujQoQCAwMBANDQ0OFQ4CsUjMBhQGTfIdQktKRYosZjILVJS7wtZU4h1iCTmBIDc5xMErQVzE2PQOTQYLyX1RkigDk1GBgyAvLJqVnnJ3FuC2gYDgv39EOxv3lFVpW/Cd3+aY6MAswKVFGtOP0CUmKwj5WxmbqIgzU2MweEFY3B4wRirIGqSuLK2waA4QaVYqgFCSnwUOrTwQ0q8ZaydnHIk1BaxpJEgeldZmSoqKvD2229bpAKYPXs23n33XYemAqAuON9FtU9g6NChePnll9GrVy/8/fffbA21v/76iwZ/U3ySwLIytM/PA+DZu9m8FbmAWCWB53K7oUjQM6ndJpaokgR6x/fowHmPl5hTRr6XNx8FAAT5m3+LcsuBiBEa7M8mqSRwt/ILpRPgu3zIOEhwd+fQYNF5S1t/CFlHypEUGy6a3JIwZXC3a+VeLDOZ8++bkvpvXDm5QfTOclk1NDRg06ZN2LBhAxobG9lUAMnJyU6pX2rrDkyK56NaWVqwYAHeeecdnD17FitWrEBYmHnnxrFjx5CUlCRzNYXifTR07oyLPfuhfX6eR+9m81YckZNHbpEiAdzHyqstFmhi8SFWGKKg5J6sYAviAgwiWwHxb1/P4M1FLFM41/3GV5S0GrMViQSMC7n3iIKXFBsuuA2fmzcJADuO+B4dcLCkknXh8TOHJw+JYBWVrCPlWDF5gM1FbfnKEemH+7fYdaTYsFCCUHvjfsRSAcyePRtR9PNLsQHVylLr1q3x6quvWh1PS0tziEAUiseh0yHs8AGgpMT1CS2bAa74NS62QJM4Iq4VhpsbCTC734a8+RMq6kyCVhAxy9j+4stYml2ABoMJocH+6Na2BY6UVQMwK0q3Rba1kIUoZ+QYUfD4CSiF8iZxx7Fi8gC2Pa5brKxKj1c2H8XS7AL0CW+DY+XVbKZxW5QTofumZNu8Pdv8lVBSUoJVq1bhyJFruc3sTAVAoQA21ob7/fff8eWXX+LMmTN49913ccMNN2Dz5s3o0qULbr31VkfLSKG4H3cltKQ4BH7tNKJAcLNdc8/jJ3/kZ/DmwrewkMWe1HEDgLYhAfjuqTtYBeZgSSUOllRaKTPEhbZgy1EkxYajtsGA2gYDu6OM3x+/xAjXmsRVEHNPVkADc6HfKn0TQmob2dp1YvXwyBykxEeid6CyeXaEldDWNq5evYp169Zh69atMJlM8Pf3x3333YeJEyfSYrcUu1Ed4J2dnY0ZM2YgKCgIx44dQ2OjOVV/bW0tPvroI4cLSKFQPA9H1uJyBVx5uZYLMesNPw1CwaVGwfOEziVBviTRJDkGmJUzrQZoFxJg4f4j1/QJbwMAMDLA9uPnUH1tBxwpPSJmBSIycBUw7rEqfRPaXAsQ5wd+iwUlk3mS2wUnNRdSiD1DalNQGI1GbN26FSkpKcjKyoLJZMLtt9+ODz74wO5UABQKQbVlKT09HQsXLsT48eORlZXFHr/lllvw4YcfOlQ4CoXimXhqlmIhZYIbYM3PaA1A1IrBtaxM6NkSWUWNSE2IZoOj+4S3waXaRlH3lU7rhyaT0aKyGj92iuv+Sx4SgbiFP7Hn1jeZWMsUgVtORcgKxK1dR5DK4E36FTpOrrOl3Akfsbgubu08pW5AbltxLa9i1apVKC4uBgB069YNs2bNosVuKQ5HtWWpuLhY0NXWsmVLXLlyxSFCUSgUz8ZTt0iL5fbhFojl5l7iLtJiW+EzcouQGN2CTR1AgqOPlFULpiMg1wXq/FjXF7EMkXlLig1n54+bcLLBYK5BpgFwT/9wi3QBUpDEmLknK6ysMmotNfzM41MGd1N0vpSVUei+kLkAoCo3UXpOIS5UVGDtR+/hxRdfdGoqAAqFoNqy1KFDB5SWlqJLly4Wxw8ePIiuXbs6TDAKheI65IJ81+0rxbLtfwG4vo3dkyxKBKF4F/4xEtPDt47xLR3ESsO3rCTFhltZlgjcfEoA4Kcxu9RqG825uYTmbdjinRbJJgN15jxMt0W2ZYO1CaQ4r61KaubeElZxE0pHAKi3Gio5X+i+iMWHSdHQ0IB43V84WpgNjcng9FQAFApBtbL00EMP4c0338T//d//QaPR4Pz58zh06BCWLFmCJ5980hkyUigUJyO34GXkFrELuqe53riI5fYhx7jBzPzgbvJ3g8HIjvXX+SOwdk8xUrIuIK2hFNOGRmLF5AFWSgxZ8C/XNkDfZIIGYMufAIDByIgqpKkJ0RYKjNS94CoY3HxRSpQofr4nsfvInxc5lARkSynXShRvfioADWgqAIprUa0szZo1CzU1NXj44YfR0NCA5ORkBAQE4LHHHkNycrIzZKRQKE5GbsFLiY9iLUtKrRqeVieLm1NpYEQY604jQdsknogoO0RZIOU+lm0/iYzcYqvxcJUQEp/EwFz+xGBkYDAxuKd/uKQSFBKos2h3aXaBxS44YhFqMJgQqDNbrUiOJaJsyM0xcUdqYM5QLhanxZ8XJdQ2GFiFz9H3WigVwKOPPoo77riDpgKguAybUgc888wzSElJwalTp8AwDKKjoxESEuJo2SgUikocmTOHy5TB3TBtaKQqWVwZBC41bn4Sx86hwaxiRGq0kRIhBAbXlYWU+Cis2JGPRkYjOB6uEhLk78cGZhPLklZjdqndFtnWans/v44buQ/k2IItR9k+iLVL32S0yLEkNx/k+nYhAWySS75ljD8WEt+lBK5sjrzXNBUAxZNQHeANmE2ier0eXbp0QWxsLFWUKBQPQSiQ1l0oqWjvKKTGTd4DwMpDZGsT7I8qfROyjpSjSt/EWoa4hXSnDO6GjKSOmDv6ZoQG++PClXpEzs9C2vpDFuNsE+wPfZOJLXNC4GYM52/v515PlKjMvSVITYiGVgOLHXyhwf4I9teyAd9SQdvc+SD/JzvwpCxGRJaF4/oqVnqIbEpr0ckhlApg6NCh+PDDD2kqAIrbUGVZIsUId+7cidraWgDmXXCjRo3Cc889h/bt2ztFSAqFogxHJAV0FHxrlVJLky3WMalxkxic+B4dWOtOakI0fp0/wmLL/cGSSrQLCWAzWwvVSFu2/S80mcxBP9xSIdxA5doGg0W8EpFPTFZyfdzCn9i0AIcXjMH+4svIOlKOgRFhitxs3Hnj1sIrvVxnUQbF1tgiR14jxtGjR2kqAIpHolhZqqmpwaRJk1BXV4f77rsPUVFRYBgGhYWF+OGHH/DHH39g48aN1MpEobgRT92lBihX5Gxx3/HHnbm3BG9mHUf9NUsPsagQqw5xb/Eze5PYI1IbLjUhGpMHdRHsMyk23EoGwBxvFHrNYgWYrVT8rfxKxiWWMFMMvnWNn89JzPXmCVRUVODTTz/FL7/8AgAICQnB1KlTcdddd0Gr1bpZOgpFhbL02WefQavVIisrC23btrV4LzU1FZMnT8batWuRkpLicCEpFIr3I6ckSCVWVEt6TiFr3alvMiHY3w/lVXr069wG5VXmkiIkIJkocEuzCyyCtIniwVWW+LvOuLFHpC0SFxXfowOyjpSzSpWcxYzftr270rgWM0+wNArR0NCATZs2YcOGDWhsbISfnx/GjBlDUwFQPA7FytKuXbvw+OOPWylKANCuXTvMnj0b33zzDVWWKBQ34Gk7zwDlMvEDsAGwSSNtaQ8wKw7EsnRPf3NeJAZmS0sbjtWHmxGb0OZaTBBRNga+sQMGoxHzrqUOEEpDQNxw3BImyUMiLKw5YhYz7ri441ZrWeIro57yHAjBTwUAAH369MGsWbNoKgCKR6I4wPvvv//GgAHiZtwBAwawfmYKheJalAZ2u7Kmm1KZyHkkwJpvSSEykyzVC7YclZU/eUgETrx+F4oXJ2HF5AFIig2HVmN2ncX36ACtBojv0cHimrmJMegcGmyRLTv3ZAWq9AbUNDL4v635iH7BOrCba8mp0jchJFAnuCPvcm2D4Pj480TGOzAizOlZ0t1R4+/vv//Gyy+/jMWLF+PChQto3749nn/+efzf//0fVZQoHotiZammpkbSLNq6dWvU1NQ4RCgKxRvwpGKySsuPcLNUO1t2pTKRnV+A5ZZ9Anc3mwawcKFJwb0/KyYPQOEis+LEtdhwFSRuWRCuAqfz06BlgAb1TSYYGeD7P8sxbPFO7C+2rJdGiuQKuc2WZhdAfy2lwPd/llvMu5DSVValx8GSSlVlSmzBlbsnr169io8++ghPP/008vLyEBAQgEmTJiE9PR133nknzZlE8WgUK0sMw8DPT/x0jUYDhuGXfaRQfBdP2qavtP6XLfW4sgvrEP92jmrFSkomriKTPCQCC8f1Fd1+TmSemxiDIH9zsG+DwSSrrBJL1Cubj4oqJ3wZSZ22y7WN7PkMw+CzcTfgn7GdoNWYcymVVemRdaTcYg5zT1awweFSkFpx3PFzZXBl3T1n9MWtdZe5twRGoxE//vijVSqADz74AFOmTEFgYKDD+qZQnIXimCWGYZCYmCiq/VNFidLccOg2fYMBKCoCoqIAnU25YgURSk4o9H8x1u0rxceHrsDEODbhoFAiRqlyGOQas6oBBOr82DZe3nwU+4svi+72YmApu2V7lrE93ASL/loNTCYGd/frBAB456E4vDdFy84pSTMwMCIMaesPsdcRuHM/NzHmWgZuI1v7TaqkiZh8jsYZuye5c7hqyy4cWH+QDdGIiIjAzJkzaSoAiteh+Ft50aJFzpSDQvE6HLbQGAy4GDsQ7U8cwcVesWh/5KAqhUkq8Jlv/SL/V+reycgtgkllRmclcBVNbnFXbi4goRxNocH+aBsSyMry8mZzCgCS84gLUVAAWGTNBsCmCOAqKyRAm6Dz06BtqyDcdlNbAJexbl8pW+7k1/kjMGzxTtYtx/+pSMZEgsfnJsYgJFDHuvu490zo/gkFg9sbxO+qTQCpCdFY/v3vCD3zK0IqT6EYgEkXiIEj7sGrqVNpKgCKV6L4G3nChAnOlINCab4UFaH9iWt1r04cMVuYevRQfLlUXiKh7eRqlB5S6iNtVE+HLrD84rbEEkHqknHzIHF3mRFlg0ASN/YJb2O1jV+sgC4AtqQHwOCm+Vnw99MgJFCHKn0TQoP90WAwQd9kRFmVHhm5RVgxOhQZuUUoq6pn5zk1IdqiMC1grgdXpW/CK5uPWihQUpY0vmxkF15tg8GiPpy95WNcUX6moaEBur9/Q2TB9VQAdR36orBNHCpr2mKhiKLkibs5KRQuNpU7oVAoDiQqChd7xQKA+VXljiCpuBNuPIzSuCYupNTHlMHdVMmkhuulPPwQoNOyQdxc5UFslxkJ3L5U28gqAtxYJqGdZaS/AJ0fyqrqAQBNJgYNBiO7S67RYGT7SImPYl+5ZUkAWMRavTG+L15K6g2tBqyipNWYLVxS94j7Hje4myhv/LIotlr4nBkLxTAMfv31Vzz55JNYt24dGhsb0adPHyxfvhwPz5iFTu3bSvbrSfF/FIoQGoYGG9mN0WjE4cOHERcX53MmZl8emxJcNn4nxSzZCxn/8Ya2rAtKTNmy1zpALCyhwf4ICdSx7Shpl3sOWXi5geydQ4MtchiRvgh+GuBaFRMEX6vtVt9kQr/ObXCptgFJUQGYN3EotFotW5YkNNgfhxeMEZSFuP/4ljA5xGLM1JQ6cbRlRu4z8Pfff2PVqlXIy8sDALRv3x6PPfYYhg0bpniHm6dbluj3oO+OX+nYPOdbmUJpzuh0qlxvgGsXGL4LSgg5N0/a+kNsRmuhYGyuy1CoPAjXSsSPaxKqQyfleiQZu4HrsU3EFVh/bYt/59DgaxaremzKb8S8a9eSuKYqfRN6vbIVLyX1tqmciRBqEksKKYjOdLHxuXr1Kr744gts27YNJpMJAQEBuO+++zBx4kSrHW5yz6onl+mhUADqhqNQvBZXui6IC0rKlcJ18wht6yfxSFlHygWv57oJha4n4+Vv2ZdqRyg1AHGhHV4wBocXjLFYpDUA7ukfbuGy6xwahAk9W7LnGEzXjfH6JpOgHPbm4BK6nn+Me/9dmW6Amwrgxx9/VJQKgLrZKN6OKmWpqakJI0eOxKlTp5wlj0P44osvMGLECPTr1w/33Xcffv/9d8nz9+/fj/vuuw/9+vXDyJEjsX79ehdJSqHYjpoFUsniK8WUwd0E4534+ZL4SR25iyM3i7YcQteT8SbFhtusGIgt2iSr9z39zVYvrrKV+3wCEqNbsOfe0/+6/MH+foJy2KscCF3PPyaVL8pW5J6Jo0eP4plnnkFGRgauXr2KiIgIvP7665g/fz5uuOEG0XZdqcxRKM5AlRvO398fjY2NHp1p9ccff8SiRYuwYMEC3HLLLfjyyy8xa9YsZGVlITzc+kv69OnTmD17Nh544AG8/fbb+OOPP7Bw4UK0bdsWiYmJbhgBhaIMNa4LITeNI1w3/DakiuGumDxANBcSYOmqIW4y7m4wR7hqxHJjcbN6y7mMxMbBl9+eHFxC14vNiSMReyaqq6uxdOlS/PrrrwCAli1bYsqUKbjrrrsUxbBQNxvF21EdszRt2jSsWrUKb7zxBnQeFIhK+OSTTzBx4kQ88MADAICXXnoJu3fvxvr16/Hcc89Znf/ll1/ixhtvxEsvvQQAiI6ORl5eHtasWaNaWTIajfIneRlkTL44NiX4yvhT4iORkVuElPhIdixCx/jwx2/ONVSElPgoTBncDSnxkVi6/S/UNjRh7Z5iNrYJYJD7fILFtQR+G4T0nFPX4qJOIff5BIu/Jw/qomicpO1buoXij9Iqqz4mD+rCtmU0Gtnz24YE4Fy1Hrd0a2PVr9JngC8/YzIhPecUGJNJ9W5CvpzkmC1zQli3rxTLtp8EAw3ib24nOD/8Z6KhoQEbN27Exo0bYTAY4OfnhzFjxmDy5Mls+Stv/2wowVe+B2zFl8evdEyqd8M9+eST+O2339CiRQvExMQgODjY4v33339fTXMOpbGxEXFxcXj33XcxevRo9vgbb7yB/Px8ZGZmWl0zdepU9OrVCy+//DJ7bPv27Xj66adx+PBh+Pv7y/ZLoukpFHvJLqzDpvwaTOjZ0sL14ykypGRdQEWdCR1a+CEjqaPVsQk9W+Kzw1fQaAKGdQ1C7w4BVm3xzyfvA7A4V24uhN4nbWtg3r7fMkCDz8aJu4fI+WQ3HF8muXuQXViHdUevAgwQ1ykQBZca2euE5spelu+twp4z9RjaJQjPDAlVdS2RB4DFeIVkYxgGJ06cwE8//YTq6moA5uzbY8eORadOneweB4XiaTh8N1zr1q091j1VWVkJo9GIdu3aWRxv3749KiqE6zVdvHgR7du3tzjWrl07GAwGVFZWomNH5V9y/fr188ltlXl5eT45NiW4evxp23NQUWdCVlEj5k0c6vT+5GSYO36wxfjTGq5bheLizBaJtAazxeJqkwlfndCj0WRWVH47U4/iq7AaT1pDKZZu/wtNDIOvT9ShSm/Cyj+u4J7YTtj70vUxx8WB3YEGWFukiJxfn6hDVlEjUuKjkDaqLTJyi3C5thH6JhOMjAZp260tKOxYG6wtUQAQUFSErl264HgDkL6rCP+MDsDc8YOh1Wot5MgqqkJNo/n3ZvFVYO9LY6za5s6VvRRvz4GJMfcVFxcnea7VfDUIW5b4spWUlODjjz/G0aPmxKDt27dHQkICHnroIUFvgpil0Jeg34O+O34yNjlUK0veUPaEH1PFMIxknJXQ+ULH5dBqtT73IBF8eWxKcNX4UxO6s7Eq9vbHjR8SKiHCP4+8T2QYGBGG2xb9DwajEfMa2mLa0Ej2H5dpQyORkVuMKr0e9U0mBPv7odFgQlJsOG6LbGsxnsy9JcjILYYGQJXenJWb8MORc/ijdJeonBm5xSirqkdGbjGmDY1k5axtMLDHUxOiUdtgNqsH+/uhvslkcQ0fofGQPErLtv+FkEAdyquvpQ649gxw5UhN6M6mIEhN6M7eMzLO1ITuormilB7joub54M+X0Fi58FMBMH5a9Bk6Gq/OmY4TJ05Ap9MJ9snvx5eh34PNd/w2pQ4wGAzYs2cPvvzyS9TU1AAAzp8/j9raWocKp5awsDBotVpcvHjR4vilS5esrEcEIavT5cuXodPpEBoa6ixRKRRBxLa627IFXelWe/4uKyLDwZJKVOkNqGlkkJFbJNkXyYodGuyPl5J6o3BRElZMHmA1HtIXYM5jNDcxBvf2N++SC/L3k5STv6OKtD03McYiA3aVvgn6JhMaDSY2k/bAiDDZuSTvN3Cydw+MCINWA8S0CxCUI3lIhGAKAv6cKtndxj22YMtRQTnV7HpTugNNKBVAQ7ubkR/1EH4x9RRMBUDglqIR6sfeFAoUiqegWlkqKyvDPffcgyeeeAKvvfYaKisrAQAff/wxlixZ4nAB1RAQEIA+ffqwOzYIe/bswYABwrtw4uLisGfPHotju3fvRt++fRXFK1EozkRqC7rcQqR0qz1RCMgONu71ocE6tAzQsO4psRQE6TmFmJsYY6U08K8hMs1NjGEXfVKy5KWk3lZykmvT1h8StbhwFQiu0kZSFQDmXW5y2/nJ+4E6LStj7skKGBng8LkGwf7kxknGIqS4iB3TaizLvdiKEsVKKBXAG2+8gWmPp6Fjx46yipZUKRryPs2vRPEFVLvh3nzzTfTt2xdbtmzB4MGD2eOjR4+2CJJ2F48++ij+/e9/o2/fvhgwYAC++uornD17FpMmTQIALFu2DOfPn8dbb70FAJg0aRK++OILLFq0CA8++CAOHTqEb7/9FsuWLXPnMCgUAOJb3QH5rf9Kt2tzt83zr588qMu1UgDdLPrkFrolx5ZmFwi6/bhySi3efHkz95awBWnLq/RgrvUvVQKF20bm3hLknqxg5xEAu/U+bf0hK9ekUAbxN7NOADDXjpNDapx8ufgpBvYXX2blWTiur11pB7jzJ6ZgVlRU4JNPPsHu3bsBmFMBTJ06FWPHjoVWq0UsrmcPl9otJJciwd4UChSKp6BaWTp48CDWr1+PgIAAi+Ph4eE4f/68wwSzlbvvvhuVlZX48MMPceHCBfTo0QMrV65E586dAZi/JM6ePcue37VrV6xcuRKLFi3CF198gY4dO+Kll17y2CB2SvNCSuFx1EIk1c66faVYseMC0hpKr8UJRWPBlqMwMrBQjgBcix3Ss4rN0uwC1trDbZ9b9oQb08QfZ3pOIetGYwALV4+SHFHE6hEa7M/2ERKoY12TxHojVjIFAAJ1ftA3GeHvJx+/qPR+8K0tZVV6nKvWs/I4Irkkvx/SHkkF8O2336KxsRF+fn5ITEzE1KlT2VQASuAqYtyae3xofiWKr6BaWWIYBiaTyer4uXPnEBIS4hCh7GXq1KmYOnWq4HuLFy+2Onbbbbdh06ZNzhaLQrHCnvpujlqIpNrJyC1CRZ0Jy7afREZuMQZGhKFV0HX3NIk/+nX+CHYsl2sboG+6/h3Bb59b9uRgSaWFVYpv6VmaXYBqfRMYgHX1kDiZYH8/NkEjACurFlFeiBLHraPGPU+KuYkxSM85haQo849Dqful9H7wlSoxeeyt/cfth2EY7NmzB2vWrGFjNPv27YtZs2YhMlJ9ULY7atFRKO5EdczS0KFD8dlnn1kcq62txXvvvYfhw4c7TDBKM8JgAE6eNL82M9wR06Em6DYlPgodWvihwWBCWZUe3/9ZzsaocAOrgesxMiT2aG5ijGBfSbHh0AAI0PlhYEQYOoeac7Xx54EET78+vq9FP8Ri1GgwoUrfhPScQqtgduIm5AeAExlvi2yraK745U6WZhewyp3aueS2ya9dxy2xQtokffGfDaV9krbv6MTg5ZdfxpIlS1BRUYH27dvj3//+N958802bFCWAli+hND9UK0svvPAC9u/fj7vvvhuNjY2YO3cuRowYgfPnz2Pu3LnOkJHiyxgMuBg7EIiJMb82M4XJ0YuOkoVUjYI2ZXA3ZCR1RKDO/FUR5O8nW4+MHAeABVuOWvW1YvIAhIcGQ99kwsGSSiuFhg+/H6HAde4xfoC0kJxK5oDM5bp9paLncNtx1M4v/o5B/pwovX9Xr15FRkYGnn76aeTl5SEgIACTJk1Ceno67rjjDrvKVjmqFh2F4i2odsPdcMMN2LJlC7KysnDs2DGYTCbcf//9uOeeexAUFOQMGSm+TFER2p84AgDm16IioEcPp3drr4vDUTg6pkPIPWKdR0k8toZ7LgAszc6HwWjEP3regD9Kq1XNV3pOIYwMoAEs6pkB1q4oNfMgdi45xo2DEkNoDvjzROYyI7cIK0aHskpTaLA/5ibGADDvIDxXrcfAiDCHuab4geZECVNy/wBzQHZ2dja++OILXL16FQDQ0O5mJE1KxpRE8dp8rsBTPncUilpUK0sHDhzAgAEDMHHiREycOJE9bjAYcODAAQwaNMihAlJ8nKgoXOwVi/Ynjphfo6Jc0q2ahS27sA5p23PYBINiuGMhyNxbwrqE5ibGCC6kQjvYpIKiuVaLKr3Z0vdHaRWbEJG0IZ9A8XrMEHGXSQVTKxmr0G47vhxK2hY6hz92kj/InDbhMjJyi1Clb0Ln0GC23+//LAcDIPdkxbX4JvsD7vmy8Z9Vbs4qcj7hv1/twI5N66CrM+eai4iIwJGWt6KEaY/PD1XicTfvW6GxThRvRbUb7uGHH2ZrBXG5evUqHn74YYcIRWlG6HRof+QgUFBgfnVRcWY17q9N+TUoq6rH0uwCSTeLO+KPSPwOVxnhu0fkcvcI5QcaGBHGBlGTPEv88XFTBgjNi1DSSH6faesPKXJdZe4tYV16/CSbauZdylXGfSa4+YNICY+U+Cir2CmyW6/BYHSaa0roWeW7/+78z2Y88e8FyPliBXR1F2HSBeLxxx/HO++8g1njEjwmvojGOlG8FZt2wwn5uquqqqyK6lIoitDpXOJ646LGsjGhZ0tkFTWitsEo+avYHTllyI4x8n8huJYIuXxNZLEftnjnNStKEFaMDkVcXDdo/Pws2hDabaY03xPpk7tlXux+EEXJyABajTlAnLtzjLvDjeuuEkLKsiFk0eHO15TB3SzKeaQmRLN5oAJ1ji0BIWQt40LGPKBzCN5f9SnaXzyMM4wRGo0G+hv64aFJU5A0orfguNyJJ8lCoahBsbI0Z84cAOZ6afPnz7fIs2Q0GlFQUCCaJZtC8WYSo1tg3sShWH/gjKQy5I6FQGmfavM1kWMp8ZEALgu2Qf7mxznxEXLXESWPlBbhxzRxIbFPWg2wcFxfUSVn2OKdsi4epQotd6xr9xRb5Jri0uZabTsSwySGWhetnLuKYRgEXDyJot9/QcdGc1xSh27d8fJzT9m8w41CoYij2A3XqlUrtGrVCgzDICQkhP27VatW6NChAx566CG8/fbbzpSVQnEr3roDSG6XFjdomJxDxqqkijx3XoT6EnKTJQ+JQEigjq3hRtyIQhDXjZCiJHSelCIkdQ/F5onkmuLXx+O66uSeCbUuWqmxFBcXY90Hb6HVyR+ha7wKY0Ar3D7hMXz83jJZRYk/Rlq7jUJRhiLL0qJFi/DKK6+gRYsWOHPmDN544w2PSUBJoTgarhVg8qAuTm1fbpFVey432JvrfpOzuCzNLkCVvom9Xmj8SmThxjGRc8WsOUoTRDrCeqYEvkJD5uLOm9uhsfECbukWKrsrTWiO5IrNKh3LlStXsG7dOmzbtg3+JhMYPy36DB2Dhf96TLLYrdgYuTv+aMA1hSKNIstSZmYm6urqAAC///476uvrnSoUheJOnB2oraZ9tedyg70JYoVy1fbJT8gIWFsmiEUEgMUiTPIucc8lCge/RpujUGs1EQrwrtI34Y/SKmQkdcQfpVUW86I0f5OcBSpzbwniFv6EuIU/CcpqNBrx448/IjU1FT/++CNMJhOGDh2K1Ss/wuJ/pypWlADrZ0Hts0GhNFcUKUudO3fG2rVrsX//fjAMg0OHDuHAgQOC/ygUZ+BKd4Gzd+yoaV/tuaHB/qwFg8xZ7skKwUK5XMiONZJ+QGmffAVKagecmCIhpQzac9/VKr1c5Yc7l+bUAda74YQQmju5+RRTcgEgLy8PTz/9NDIyMnD16lVERETgjTfewPz589GxY0dF4+LCL5osVkSZy7p9pUjJuiCZnJNC8XUUueH+/e9/Y8GCBfjoo4+g0WjYYG8+Go0GJ06ccKiAFArg2vwsXBeIVMV1R7TvjHOJK4jkNwoN9lcUy8N1GwHA/uLLIHXR4uIgmUeowWC0cFFJ7eASCiSXUiSE7rsSl6A9uxP5z8Dhw5cxZfD1HYHkHKnrlCK0o/HChQtYs2YN9uzZAwBo2bIlpk6dirFjx0KrVbfzjjtX/DlRMkfcmC1+gDuF0lzQMAzDyJ9mpra2FgMHDsS2bdvQrl07wXNatWrlMOG8BfOX6WHExcWp/iLzdDxlbO7K/OuK8ds7Nv71ZFdYaLA/QgJ1qtsl15PcTB1a+GHvS2MExy+kmJE+ASgel1i8ldjcEBk7hwZLVr2XQ8ncc5+B+Ld3qe5XjawNDQ349ttvsXHjRjQ2NlqkAphxLRWAWpT2LzYX5t2A+Ugb1VOxsuTMz6s7vgs85XvQXfjy+JWOTVVSypCQEHz++efo0qWLxW64Vq1aoby8HCtWrLBbcApFCG/diaYEe5Mq8q8nbp+5iTE2zRm53lx7LQgTerYUPZfvdgOuxyo5It5K7L6LubbE3HZixx25S82eaxiGwe7du/HEE0/gyy+/RGNjI/r27YvK2Ck42fZ2rNl/TnIcjpBZbC5IfUAlOyPl2nIEzo4ppFCEUJ2U8rbbbmP/X1NTgx9++AEbNmzA0aNHERMjnWuEQvEEXPHLVE0fatxFQm4p/vVKXUFiMlq7oA7j6a8O48e8c0iKDceKyeL51OJ7dLDY2aZ0XHxXVNr6Q8g6Ui7an9gYheaHm9SS785T66qzxc0md01xcTFWrVqFo0fN5Wg6dOiARx99FMOGDcMX+0ot5LPFHc3dFcn9m48jk6o6M0GrO5K/Uig21ZbYv38/NmzYgJ9++gkNDQ2YMWMGli1bhogI3/vVT/E9XBH/pLaP2gaDRZA0F6mYE3K+2nFIKRBC/HDkHBgA3/9ZLqi8kPECsHD1qF3Ql2YXYGl2Aar1TWx/anbLCc0PN6klf4F1Z0bpK1eu4IsvvkB2djZMJhMCAgJw3333YeLEiewON758tioKSp5HR86FM+fVnfeM0nxR7Ia7cOECMjIyMHr0aDz77LMICwvD2rVr4efnh/Hjx1NFiSKKpyW+c8V2aTXuGqndUOR9/jZ8excLrgJByoRI3Z8gfz+LVz5Sc6r0/nPnQednLqmk89PYvKONoDSppaswGo3IyspCSkoKtm7dCpPJhGHDhuHDDz/ElClTJFMB2Hr/bd3hmbm3BPFv5yC7sE7VdRSKr6FYWRo5ciQKCwvx6quvIjc3Fy+88AL69evnTNkoPoKnxRiIbZd2pFKnZlEj29SD/f3Ysh/89x2dyoCrQBwsqZS9Py/e1ROdQ4PxUpJwkDF3TvnzyL3/coVsyXb9kECz0TskUGcxdlvukdC9cLUCT/r775fb8fTTT+Ojjz5CTU0NIiIi8Oabb2LevHk2pQJQipLnUTwerh6b8mucJhuF4g0oVpbCw8Nx8OBB/P777yguLnamTBQfw9l5i9RC5OFbVNyl1CUPicDhBWPQNiRQ0LokV05EDeR6ABY5heTuz5TB3QQXW9LewIgwi4SOQgHnQu8JzcPhBWMs8j5x+3XEPSIuyLIqPRZsOWr3XCq5/qNth6DN+w45695DSUkJWrVqhZSUFLzzzjse86NTaG7N9046yJ9CaQ4oVpays7Px9ttv48KFC7j//vtx33334dNPPwVgzq9EoYjhaTvZiDx8i4q7lTol/durLDhaISTtHSypFFW+uBm8lZb9EFMQld4jKUWGuCABsDFb3PPVuA3l5rKhoQHz3/oQbQ59htCrRWCgwdUO/VDScwout+3jkm3YSscjNLfJQyKQ+3wCEqNbOFtMCsWjUZU6YODAgVi0aBF2796NSZMmYevWrTAajfjPf/6Dr7/+GpcvX3aWnBSKwxFb1N2l1Cnp316FTuh6tQqUnPIiNg5u2Q8Aiq0ytsRsSY2JyHxv/3BBi5fS+ZC6FyQVQGpqKo7v3gY/xoj6lp1R3X8KKrsNx5laOExhlUPpeNz9/NuDp8VFUnwPVUkphSgsLMSGDRuwZcsWVFdX49ixY46SzWugCbt8F08dvyPTH4gVf03PKURKfCROnzmDrKJGpCZ0R/KQCMQt/IlNQHl4wRjF7XL/Jgu4WKJE7rmAecHnFtxVkuRSzfwI9UeuVfIMcK8fdoPJIhVAi9ZhON9pKB6bmIhpt9/ksHuntB17+/PUzwAXRyUpFcMb5sCZ+PL4lY7NbmWJYDAYsHPnTowZI/7l6avQB8l38dTxO3txuN5+EBobG1FRZ2L7IsqSBsA9/cNFFRgpGeUWcKFrnT1mMZQ8A8MW78S5i5XoVnkQLSuOQQMGAQEBmDhxIu677z7RHW72KDKumg/u+NcfOOOWTPpyODt3mqd+D7gKXx6/UzJ4S6HT6ZqlokSh2IsSFwL/HLJzTGj3nCP6IrFFKfFRmNCzJTqHBrFWl7mJMdAAbB4kOXeXkJtKzuUjdK27Y8rEMBqNGNWyDL2KvkSriqPQgEFDu5vx4YcfYvLkyZKpAOyJIXP0fJBnI239Icl4L1vldaarzJtdiBTvwGHKEoVCUQdZPJZmF8guQPxFKnlIBEICdaK5mewp78GNLZoyuBsSo1sg9/kEdiFKHhKBIH/zLzCdn0a1QiS1aArt1pNrz5WkrT+E6BeykLb+EADgyJEjePrpp3Hop2/gZ2hAm47hqO4zEdMeT7NKBSA0bnsC1h09H+TZyDpimwKstH1PSSFCoaiBKksUigLW7St1+K9ibtZruQVIraVFbGFSstgpOSdQZ/7qCAnUWbjJ1AZsq3kPEFYaXBncm3WkHEYG+On3AixevBgvv/yyRSqATz/6ALsWPSJbioXgiIB1R2FZE9A2i6CS9j3NMkihKIEqSxSKAjJyixyyWAntJFNS8FZokZJKtjgwIkzQTadksRM6h68scvMgAeoWc6lFUywHFkGoH2coEmIK2F2926NTxQHEFH2NPXv2wM/PD90H3omSnlNlUwEoURbE+lV6bdzCnxC38CdFiiO/L3LfV0we4BQLnhpFi+5uo3gaditLNTU12LFjBwoLqWmV4rukxEc55FexM0qXCLWfdaQcAITddAYDcPKk+VUAoYWKryzyZVdjNZAat1gOLII9sUxqFmC+AsYwDH799VfU/e8jdKz4A4zRgH79+mH58uXYFzgQZ2oZh2zNF1P8lLg0pcrmiGfn9ky3mCfLRmmeqFaW/vWvfyEzMxMAUF9fj4kTJ+Lpp5/Gvffei+zsbIcLSKF4AmIZrNUit7CLuZmUWgxSE6Kh1YBNumhlXTIYcDF2IBATY37lKExiMVTZhXWyySQdrfiJzZNSCxt/TESZsMX6VVxcjM8++wxLly7FxYsX0aFDB8ybNw9vvPEGIiMjBWW11TKi1lXFHRO3XAz/evHs3OJWPHdCXXYUT0On9oLff/8dqampAIDt27eDYRgcOHAAmzZtQnp6OhITEx0uJIXiK8hVTBeqDk8sBuT/5FUofxH3Gn6yxeQhEUBREdqfOAIA5teiIqBHD4u+Q4P9LRaqTfk1qNKbUweIKSSO3rbtqMryfGWCm0tJrv97e4fhiy++wLPbtgEMA63OHw89+AAmTJhgscNNSFah+6gE7v3j/i0Gd0xScyY0dnI+SUGgVlZnIjUWZ6cJoFCEUG1Zunr1Ktq0aQMA+OWXXzBmzBgEBwcjISEBJSWe88uEQnEkjg7wVhObwrUYDIwIY+uaLc0ukN1NZ9VeVBQu9ooFAPNrVJTVufwYKn7qAD6ujBlSez53/EqtX0ajET/88ANSUlKwdetWgGFQ1ToKF/snY9KkSZKpAIT6VSqzmt2RBG4pGan5khq7M604jow9smV+KBRHoVpZuvHGG3Ho0CHU1dXhl19+wbBhwwAAV65cQUBAgMMFpFA8AaEAb3sWAjWxKdwCswdLKmFkAO21coxCu+mE4qIAmF15b+5E9sqNQEEB2h85COh0Vn0DlgsvP3UAF5KTKdjfT3XOJ1vmR+35UkHwQrKSVAArV65ETU0NbrrpJgyfPAf6HomYPTpOkSxp6w9hwZajGBgRpio+ibs7ku8+ldsFaI/C6syUDLbsbpRrCxBwL1MoTka1svTwww/j+eefx/Dhw9GxY0cMHjwYAHDgwAH0uGbOp1B8DaEAb7ULlC0FYfmQ6xaO62uxI00u2Job/Pvh7hKz600n7IVXMy7SbqPBJJrziT92NeNUOj9qzhca3/nz5wVTASxfvhxPPzgSGUkdMWVwN0WykPQCJMheSmbuos+17PFzaMntAuSO35N2ksndF1viyITmh0JxNqpjlqZOnYr+/fvj7NmzGDp0KPz8zPpW165d8cwzzzhcQArFE5gyuBumDY20OCYXA8OPreAuDLb8ks/cW4Kl2QWy5wnFe6QmRLPXyikUamJ7yLncum1iMVRq4mLUxiypOZ87voaGBnz77bfYuHEjGhsb4efnh7Fjx2Lq1Klo1aoVALNbjqAkXiYpNhxZR8qRFBsuK7PUM8G9B0L3RCxeyZNikOTui9JnTSouj0JxBaprw73//vuYMWMGgoODLY7X19fj448/xpw5cxwqoDdA6+b4LkajEUu+3WNRSFYp3NpdXGVlbmKMlVtIScAqaQ8wu90AOLQ2mHDh21NIigrAvIlDFd9/IqdWAywc11c0CN1e+eyBYRjs3r0bn3zyCS5evAgAaGrdBUkPJmPOvUMtzuV+Bga+8bOiIsJK4Cq//GfC3nYdGQDtCd8B7qoLSPCEOXAnvjx+p9WG++CDD1BXV2d1XK/X44MPPlDbHIXi8WzKr0FZVb1qkz/XzbI0u4AtISK1c4ogVKcrNSEawf5+0AAYGBHm8PxCwjE09ViXdxXxb+eocqGR9AVieZlswVGB5MXFxXjxxRfx9ttv4+LFi2jROgxXeyThROe7sf54vV1tq4G4MK/WN2F/8WXV9QHF8ISyMI6GphKguBvVyhLDMNBoNFbH8/Pz2V1y7uDMmTN48cUXMWLECMTGxmLUqFFYsWIFGhsbJa+bP38+YmJiLP49+OCDLpKa4onwFyWyG0xtPhpu/TbAMgibmztpYESYaDzUd5xCtclDItA2JBAMrsfD2FMqQ6g4L1cO899BgAaqlMXkIRFYOK6vwxc3vnxqY3OuXLmCDz/8EM888wyOHTsGxk+Hcx1uxeleUzHj/rHoHNZCVN7swjrEv52D+B4dLDKX2zseolRK1WMjNOdEjUoUQE+K1aL4HopjlgYNGgSNRgONRoPExEQLhcloNKKurg6TJk1yipBKKCoqAsMweO211xAREYGTJ0/ilVdegV6vx7x58ySvvfPOO7Fo0SL2b39/f2eLS3EgjnY7ECvQ0uwCTB7UBYnRLTBv4lDEv71LdSwIP66EwM2ddLCk0sq1kJoQjQVbjrI737gKDDnOlUNqDsTiQvhxRPz4kuQhEZg8qAvHDalc8XFUniSpNrn3SWoejEYjtm7dinXr1qGmpgYAcMcdd6BN/9H4/FClaI4i0lZKfCQ25degos4EwPpe2TMeAFYxX2KoiSVrjtia24pCUYJiZenFF18EwzB48cUX8dRTT7HBj4BZuejcuTMGDBjgFCGVEB8fj/j4ePbvrl27ori4GOvXr5dVlgICAtChQwe7ZeAGgvoKZEyePLb0nFPXLB+nMHlQFwe0yLCv3PGnxEciI7cIKfGRiudj8qAurEzca1LiI7Fs+0kw0CAlPhJr9xRfazsKUwZ3w+RBXcCYTOyxyYO6wGg08o5fl0NqDqRkkBuP0WhEYnQLzB0/GFqt1uHPwbp9pRbjVof1feLPw5EjR7Ds/QxcqTgLALjpppswY8YM9O3bFwAwc5S5BaFxsW3tKsKEni3xQ2Gj1Vxx5Qegeizce0MQm2Ox++hsSF+Ze//Gyl/+tvFeWWPPvRe61pbPp1K84XvQmfjy+JWOSXWA9/79+zFgwACvsL4sX74cv/zyCzZu3Ch6zvz587Fjxw74+/ujdevWGDRoEJ555hm0a9dOcT8kQIziHrIL67ApvwYTerZEYnQLh7YHwKFti5GSdQEVdSZ0aOGHCT1bivYpNla5ObD1fUfMrVQb3HFnJHVU1B85J6ZdAAouNVqcS967q4sRDQW/4MSJEwAAgzYQtZ1vw9JHEtgdvHKyAvL3nis/ANF76OhnVExmW9pWej3/XtmLmvb4MjpaFgpFLsBbkbJUU1ODli1bsv+XgpznbkpLSzFhwgTMnz8fDzzwgOh5P/74I1q0aIHw8HCcOXMG7777LoxGIzZu3Kg4ySZRlvr16+eTOwXy8vKsx2YwmEtlREWJ5utxJPZZIGzv4863clBeXY/wNkH45d8Jstcv3f4XNGDw3OgesnJy+9v/92X8mHcOd/frhD9Kq1BWVY/OoUHIfd6yz/i3c0Tfk2rfnFRT/DqxdtWMXwwpmflzrmR8UueQVABbtmxhUwFExQ3Dfm1vPD6yt+w9EWqbfAaO1odaWVbELEv8+VZz39Rib9ty10uNXwwln1c159Q2GFClN7AyuuL7gIvo92AzwZfHT8YmpywpWuUGDRqE3bt3o127drj11lsFA7xJ4Df5Jeco3nvvPbz//vuS52zYsAH9+vVj/z5//jxmzpyJsWPHSipKAHD33Xez/+/Rowf69u2LESNGICcnB2PGqNsarNVqfe5BIliMzWDAxQG3of2JI7jYK9YqE7QzyMgtRllVPTJyi63yHTmzj9ThUVixIx+pw6Nk721GbjGqr8UhSclJYmHMC0ATMnKLAZgDfX/56xIAXCtt0hbxb++yiL9JTejOxq0okYeMR+46sffVjF8Mqb6nDY20mCcl4xM6RzwVwDTMufd2xbIOjGiLc9XlGBjRFusPnGFjlnoHAit/+dvq+SDyX4+VMqeX0Pj5Wcio9L7ZEn+n5pmQG7PU9clDbsIjw5TFSyn5vPLvvVQ73HqFWq1W0bXOwJe/45XQnMevaIX77LPP2J1un3/+uVMF4jN16lQLhUaILl2u+/zPnz+Phx9+GHFxcXj99ddV99exY0eEh4fj77//Vn1ts0GiGKuzcEVwq1AfUwZ3Q+/Ay4iLk//1ys2lRHbOCS16JDA52N+PXQD2F1/GuWo9GgxG6JvMRWsPllRaBazamnxR7jqx99WMX23btp7LP6eoqAirVq3CsWPHAJg/w4Vhg1Ds1xlXjusx515lcmbuLbHIvp17suKaMluEFaNDcUu3UJyrPoeBEWFW1woFy5PjSscl1I4S7A2mJyV0DpZUWhznBrj3li+HZ4HU51VIISTHSKA795W0R4O2Ke5EkbJ02223Cf7fFbRt2xZt27ZVdC5RlPr06YNFixZJxiaIUVlZibNnz6JjR+oHF+VaMVbWssQpxupN8L+07V10pLIoc/tqMFwPKCQ7q9JzCmG85hDXaszK1m2RbdlrlFochMbkq1y5cgWZmZn46aefYDKZEBAQgPvvvx8TJkzAN4fOqVauufeAvHYODUZKfCSAy/ijtEpQqQCElQOhvFVy94/bjqN3eSrpkwuRnyiLapB69oQUQnLsXLUeRgbsKwBVuw9dNWeU5odqbSI/P1/wX0FBAf7++2/ZvEbO4vz585g2bRo6deqEefPm4fLly6ioqEBFRYXFeWPHjsX27dsBALW1tViyZAkOHTqEM2fOYN++fUhNTUVYWBhGjRrljmF4Bzqd2fUmUIzVWSjNFSSE2Dn2tCnXF8mdRCxM3ErpgTqzGbu+yWSV4yhQp2UtG/uLL7PtkuuXZhdIysc9z1fgj9doNOKHH37A448/jm3btsFkMuHOO+9Eeno6Jk2ahMDAQNGiwFKQe3Bv/3A2l1JqQjQycouwfG8VahsMCA32F1TAhPrj5oVSmiOJm0+IXPPK5qOIW/iT4udR7fMrlsOIyE/isRyFUIJJciwpNtziVa0luTnnoqI4F9Wr3Pjx4wVjltgGdTrcfffdeO211xAYqNJ2awe//vorSkpKUFJSYpFCAAAKCq4vHMXFxbh69SoAs//15MmT2Lx5M65evYoOHTpg8ODBWL58uccEqnssOp3TXW9clOYKEkLsHCVtqk1FQK4FzL+I4xb+xOZT4i60/FxJ5F/m3hL2PeIS4n/xOyOfjCf/IueOt19wFT7++GOUlJgVgcjISMyaNYtNBSB1rVr3HkAshPU4Ww2YGLOlSaodpbXelMDNqcXPJyWFo54PMh/mDSyX5S9Q2a7cMVuguagozkK1svT+++9j6dKlmDFjBmJjY8EwDPLy8vDJJ59gzpw5MBgMWLZsGd555x3Z/EaO5L777sN9990nex5XcQoKCsLq1audKRbFQYh9mSr5chQ7x542+ZA6Xw0Go6j1gV/qhNsHV1lZOK6vYKJCbjwUkZPbf3pOIeJ7dJBNbsiXmxts7okJ/VITovHRtj/Q79I+vPLKYQBAq1atkJycjDFjxkgGnNq7eJqvP4XIVkDxVduLEHOVYbFYNj7k/Vc2H2UzSimX2X0Kg1rF25GKuq+7ninuQ7WylJGRgZdeegl33nkne6xnz57o1KkT3n33XWzYsAEtWrTA4sWLXaosUZofSr9k1X6Bcs9fu6cYK3ZcQFpDqcUOJ6H6bsSC1DYkkH1/bmIMq0TVNhhw73u7caSsGgBwb/9wq5gNIYsEt4+yKr1gxm++RUsp5DrubiNPoqGhAZrC3bjh6CacuZYK4K677sKUKVMsEuOK4Yg4tMmDuiguIirXn5DVR+o55j4fSu+NrWN2lNKidow08zbFG1Ads3Ty5EmEh4dbHQ8PD8fJkycBmJUnfqwQheIIuPEYzo5PyNxbgv98fxwVdeaM2VL9kaK5fKtS8pAIHF4wBm1DAlGlb2IVJeB6fTdyPV9Zkavdxu/fFmWHXBffwzKDPel73b5SVe0JtWFL7BfDMPjll1+QmpqKr776Co2NjYiNjcW7776Lxx9/XJGi5IkI3Se551hJXTRH4KjPk9ox2vrs2gKtH0exFdWWpaioKKxatQqvvfYam7SxqakJq1atQtS1XVHnz59XlQGbQlEK90vX2e4GsjvKTwOkxEdZWJb48H/Np60/hO//LEeQvxYvJfViZW0XEsAqTEmx4aLX88cqt1vPVmsCuY6/e8+enVBi8kvBtTwM7Wi0SgXw2GOP4fbbb5eMl3QXfKuJnKVIaLs8IO/iU9q/2usGRoQh92SFpAtZDULPotRn1ZWuM2rFotiKamXp1VdfRWpqKoYPH46YmBhoNBoUFBTAaDTio48+AgCcPn0aU6ZMcbiwFIqa3EGO6esUkqICMGVwN2i1WsH+hBarrCPlYADom4ySrjUpBkaEobxKj8u1DcjcW6JK4VAqp+VYry9m5G+ybd4W1OTaSc8pxLmLl/H56p+xoeKYYCqA55f8zyKOyxbFwBYX07p9pawrVigRIn8BJn8vzS4Q7ZMbyA8oc52KjcFWpZS/XR+wdCE7Ens+q46MaXJ3PBfFe1Hthrvllluwc+dOpKWlISYmBjfffDPS0tLw888/Iy4uDoB5x9zMmTMdLSuF4jKXBKGmwYh1eVdZd5SQGZ9s13+Zs8U7KTYcGgDB/lqbv5gPllReU7hMilwjcm4Uqff580r+tqWUBJkjALLxV+k5hTAajRgVcga9Cr9C0Pk8wVQA5PysI+U2uYpsdTHxXbFCEDcSSRPRLiQAWg3QYDCJpqYgipJWo9yiJDYGOTeWmOuau11fyIXsKTjS3e7q7w+K72BTgpyQkBBMnjzZ0bJQKB5Fek4hp3xJEaYNjZT9FU92lKUmRFtZQNT+Qk5NuJ4RXMkiJver2dZf1cv3VuG3DduQFBuOFZMHyJ6vxNIxMCIM56r1aNdwFvdMnQld3SX4QTwVAJGdv0NQaE5JEH1s5zb47qk77Bo73xUrBN+VSSw1gTo/tA0JEExNQRSlheP6Kl64uXNAcnmRuSCWKaH5EHNdO9Iyy3fpAebNDUrbV2P1VHMtheIobFKWSktL8dlnn6GwsBAajQbR0dF4+OGH0a2b8wsaUiiuIjUhGm9nF8BoMCAlPgqZe0tQ22BAsL8fahsMrGuMu+MtUKfFwIgwqzxKgPKMztyFJyRQp3gRkFv8bF0c95yph+la3ic5ZYnMkZyV4lDB3+hSmgNcLYYOgEkXhCdnPSqaCkBMdiHFjMSEcYPpbR073xUrf26hrKuQr7AQuIs+GRtf+fh1/ggrpUzsGSPHpBQkR++A47r01MQF2RNLROOQKK5AtbJEdqj06tULt9xyCxiGwaFDh5CUlISMjAwMGzbMGXJSKC4neUgEGJMJK3bkA7ieHkADs2uMJAnkL0DDFu8UdLEQS1Ftg4GtDyf0Bc9feNy9CAztEoTfztQjKTZcdnElc8RN3si9xtDYgMwvv0FY+UHAZAA0ftDf0A+TJk/GXf/orVo2IatDbOc2rGXJXripA4RQUl5GaVkd8kwszS5ASKBOUPkAwCqjQjm1hOZDqD81+bXkYra4/XKVOzWWICnrkZwyROOQKK5AtbK0bNkyTJ8+HXPnzrU4vnTpUixdupQqSxS3Y2ugs9DxjNwiNl6FVHe/XNsIfZPRql0C/8s7buFPAMyWAbIISuU1EnM5uYtnhoTis2s5hvi75vgILVzpOYUoq6zD6g0/osXfv6BFYw0AIDY2FjfeloR1x/UIDA5hz1d7//jnENebHNxEooE6rSq3kcXYbMwgL4WY8kGUUa0GuC2yrZWlT6kFjcjEfQ7F5p37GRBTltRY7oTmQ+p6OWWIXKsm4SeFohbVAd6FhYW4//77rY5PnDgRp06dcohQFIpa1ORfEntf6HhKfBQCtUBZVT3ezDqO1IRovJTUi60dJiQHf5dXlb7JIpaJXCsWaEqCUFdMHiAbjOrqvDFiwcREDm49O8KkXoHoVZaFVie3QttYA2NgKwy9bwZef/11rDp01aqWHf8+8McodJ/UzgMJsq7SN0HfZGItOkrrDKatP2RRA5A7H2ryY3GZmxjDPhvcZ+DwgjE4vGAMkodEIDUhGhqYC/3aU/9P6DkU+1ykxEehQws/h9WIU5tXSWlQtiMDwSkUPqqVpbZt2+LEiRNWx0+cOEFzK1HsxtbFX2iXj1Sgs9D7QsenDO6GJpP5/2RXGvnyBqyLtPK/sFMTzMkqSZwTIL5DzBZcvUCQBTs9p1Bw3GS32iubj6LPC5vwj8dewI41b8H/ShkCAgIwZcoUbMxcg/nTx4nmTOLfB6E5VZvYkQ8JsjbvWPRDaLA/AMi2wR8nN5s6X1knyheg7J7LKQWk7SB/1V/bivoaGBEGreZ6OR1n4awdaWqVMApFDao/dQ888ABeffVVrFy5Er///jt+//13rFy5EgsWLMCDDz7oDBkpzQhbF3/uF6Xcl7HY+2LHh3YJYhdVuQWa/4WdPMQyg7facckpj+5YIKTGnRQbDi1MaHv5KCJOfIFWF4+BYRirVACEuYkxrKJCxsi/D/xFXEhZFZoHqbkjbd7TPxwnXr8LhxeMYS07UnPJHSfXfbVgy1ErZR2wVr6U/hiQSlEBaCQtm7ZaGg+WVMLImF+5cN1warHX8qnmepoWgOJMVMcsPfnkk2jZsiXWrFmD//73vwDMGXbnzJmDRx55xOECUpoXtgZrOmIbtNBupFu6tUHBpUa8dm9vq3gNpcG0YucqQS7exZHbv5UiNe4///wTl7f/gCsXzwIAGoPbI3HiVDzz4EjBtvjuH6GxiC3ifGsiCZxX0q5Qm3JzuW5fKTJyi61iYvgB/dwYGsH4LQXxS1LnBer8LJJYCiWaVBMfJZRJnNtmSnwUVuzIt8kNp0YeoUzoQrtKKRR3oFpZ0mg0mD59OqZPn46aGnOgZsuWLVFXV4cDBw5g0KBBDheS0nywJVjTlu3P3AXiYEklBkaE4fs/zVm3yYJbpW9CeZUeDCAY3KpGUZHakSQltyfu9BEay7lz57Bw2QcoK/gTANCqVSskJyezqQDsyaMj9j73OIkNA4AFW47KtmvLvGbkFqGsqt5q4eZvzeeOlZ+ZW2m//POIZSU02N/KosRXGtWOi1wPwMJiR9rMfX44egdeRlxcN1YW/o8KofuqNI2E0DiIsqs2cSeF4iw0DMMwjmgoPz8fEyZMEIxn8nWMRqPiquTehjPHJrWAki/rzqHBsqUg1JzLv0arAfuFTLZoB/v7ob7JBAZAkL8fWvkDQ7p3xB+l1bILhBpskdvVyN3/+vp6bNiwAZs2bUJTUxMYaFDfqT82Lnveotits8dKdrZV65vAAA7th8zB8Ya2yMgtlsyjxLWGhAb7q8qTJYXU/NmbK0noeu4xkjohjrcjkutqFJJL7T1XU2PP1fjyd7wSfHn8Ssdmf6QghWIjUvFJUrE4ancbCcU9kMDrAJ05uJeUfAj217KKklYDvHRXT2QkdcQfpVWsrI4KqvbmgFSGYZCbm4snnngCX3/9NZqamtAxogeq+0/FI4/NYBUlMvdCu8bsgX9PSWzY6+P7qopdEntf6NiUwd3w6/wRyD1ZYbWDj8C1hgDiMUtkN52SWBw5C40tsTrc8QkF7Uu1yX1upZ5he3e90RgkiidhUwZvCsURSLkMpFxcfGVF7teoWF4X4rppGxLI5qvhWpwWjut77Vf1ZaTER7HxKtw+7cEd8UaOoLCwEKtWrcLx48cBmGMWH3vsMdx+++1WO9z4Lh57c+HIJVMUmlNuskcik9zzwT1mTkwqnZSRwH+mxWKW1CQcJc8pN9GnvfA/Q8QaRlyYUv0IxWRJxdPZes89ybJEoVDLEsVt2PrLkfuLVUhxktuhRr68ScFT7lZpci6/ZhexKpAFQExuV+c9ciVXrlzBhx9+iGeffRbHjx9nUwF88MEHGDp0qGAqALE0AEpyGgnBVb6E7qlce0qeD8D8TGgAXK5twJtb81FRZ8LS7X8BsMyHxIf7bAg9J0K76fjYmqdJqh2pNok1DACrwAmxbl+pTfmubLXEOsqCS6E4AsWWpZ9//lny/TNnztgtDIUiRtr6Q8g6Us4Wc+UuQNxf79xf9mKBtvxf90p3RCnJLO2Lu3eMRiN++OEHfPnll6itrQUA3HnnnZg+fTo6dOggOS986wJRTGsbDFZKrlxNNUC8rprUrqu5iTGS1h6he36wpBIMzLm1iAqoASN6vlKUBPrzx2Jrf1zF9Gp9k8VzyW8zPacQ7UICcKy8WjTPEj/AXcgyLHQfbN2k4ImbGyjNF8XK0pNPPil7jliSOQpFCiXm9qwj5TAy1sVc+V/63L/FSnOkJlyv0WZiGMVJ+OS2Qfva7p3MvSX4ePNO3FCWC331JQBAZGQkZs2ahb59+7LnkXl5M+sEFmw5yiq0XITccfzt7kpcU7akZhB6RuQgzwgA3HlzO/zvxHkw0LAxPo5AzKXoKCWBtFPbYJB8Lsn8xC38CUYGbHkVPnxXtNC9GBgRhnPVeovPlK3Knre6qSm+iWJlKT8/35lyUJoxSnKxJMWGs5YlIYQULqEvbsAyXgmwzt8jhpot7t7+JX/u3DmszXgHrS8XQg8gIDgEMx99BKNHj7baMULGTdIs8BVa7jliCy2xLOWerEBtg0G1UmLLwqrEIgaYLWtD3vwJFXWWMVJqYmqkcgjx6wQ6QklQs71f7Jkm76XER6J3oNkVLRezJZYTi0LxdmiAN8VtCCXDE2PF5AHsAqw0eZ3UFzfXcqBmt46SwFcleGrwKjcVQOC1VACX2vaBX/dhGDt2rOA1ZNxcVykX7nGpQGBA3BroDLiB31JKxORBXTChZ0tkFTXKupzE4J+7NLuALbciVcCXpESAzHl8ubnWKqm4QK5cxF05MCIMwxbvZNv4z/fHMWNAaxxvEE7KyYW6zii+Cg3wprgN8kVN6mtJLQTcwFGhoG4hN4NUYGzykAiLAqWuxtOCV9f+9jf+8dyHeGTGbDYVQP/+/ZE4499oiroTj4/oLdvGiskDULgoCbdFtrUI8uW6UIXg3ltyz8iCTdpwROC8mjb49ycxugVyn0+wcueGBvuzljApxJ7FNsH+kooSKfZbpW/Cgi1HFQewA9cD4KXSFXDlIgHpB0sq2TZI/rFN+TXXYpakn1lbN21QKJ4OVZYobkPNLh/u4sW/TmwHm+QXt8EAnDxpfnUDnpRjqbCwEGvfW4RWf22D/moVOnbsiBdeeAGvvfYaUv85BBlJHTFlcDfF7fEVjaTYcGgABOi0gos93+rCXbD5uxzFdtEpUYRIbTWupUZoVxs3r9HAiDDEv52D7MI6q/aSh0QgJFCnqOYf/1mU2lHHnRdifdJAeqcaP5/V3MQYtj8yd9/9WW6l7HDlEmpj4bi+6BwahAk9WyIlPspjnlkKxdVQNxzFbahxW3GDsvcXX7a5HQCAwYCLsQPR/sQRXOwVi/ZHDiLz9zKXusU8IXi1uroamZmZ+OmnnxDEMDBpdLhy40B88+5ci2K3cvBdinxXzIrJAyyUH/64hVw3/GPcYGWhHFtqXGIEsXtA4tk6hwZfk7sem/IbMU+gDVvdTkruP7ft/cWXkXWkXHQzAj+Ant8OcVNLbT4Qa+N6Bm/5mCUCv5yQp7mbKRS1ONSy5KDKKRSKFdxf8VlHrH8h85G0NBQVof2JIwBgfi0q8ji3mDMxGAz4/vvvkZKSguzsbDAMg669b0H1LY9g9vRkK0Vp+d4qdH9pG3q9slXSMvRm1nFEv5CFrw+ctjpHziVKlOG4hT+xwd1C2ZyJRYavIClxiSmx5vBlNf/fbFkRQq3bSY0rkNt27skKyZ1qcvNrthBZW1/F2rDX7UnujZLPKoXiDai2LH300Ud4/PHHrY4bjUY8//zz+O9//+sQwSgUPmSB5P5aFYPrtrGyGEVFmS1KxLIUFYXUhEDFFgJPDc6Wgsh8X0QTinZ/h9OnzQpNZGQkZs+ejT59+oheu+dMPZtzSMoyRHbDHSmrBgAr649UfTDu7kQ1qQOk8hPxxy4ng1gfxLJiL87Mw+WIzQdigfaTB3VRLY+azyqF4g2otix99tln+OqrryyOGY1GPPPMM82yiC7FPtT+0k5NiFZk1ie/kgHr+lzQ6dD+yEGgoMD8qtOpshB4oxXqo21/QJe3GbvWf4DTp0+jVatWeOKJJ/Df//5XUlECgKFdgqCBucCwmOXi1/kj0K9zGwBA59Ag0QzrfLhxMqHB/myckJL6afx7JmZd4crAf95cmXFdSR4usTp1gDnFgJxVTKodNefbG1NH7s2KyQOsPle+nOWe4ruoVpZWrlyJpUuX4scffwQANDU14V//+hdOnTqFzz//3OECUnwbtYqH0vOF3DYW6HRAjx7mV5V4UnC2HPX19cjMzES7P9eizdW/odH44Z///Cc++ugjjB07VlEF8WeGhOLUm2Nx4vW7JJXJS7WN1/6nYRdIubki9zP3ZAVCAnWYmxjDxjd9fy0gWahgrRi1DQa8mXWcdecB6svjOAviKmwV5C96jpA8xOoWEqhTbI2SGpeSkiTO3NXmjT82KBTVylLfvn3x/vvv45VXXsGOHTvw1FNPobi4GJ9//jk6dOjgDBkpPozQYpq5twRxC3+yWPCkzpfCGV/6trRp7y99tTAMg/9b/S0mJj+Gr7/+GiajAf3798eKFe9i9uzZaNlSOAaHz7p9pUjJuoB1+0plzxW6N3JzJWQBJMeC/M2KXLW+SXIeyFyRvEn6JhO7Q00o+FxoJ6Utiq/UlnwxrtY3Se6eE5LHFhmlrhFSVuz9AaDmefWmHxsUCkHD2BiV/fPPPyMtLQ1RUVH47LPP0LZtW0fL5jUYjcZru0XiFP1S9ybcMTYSLwGYc8UoiTNxFo4aPxmT0vGoPZ9LYWEhVq1ahePHjwMAjIGt8fKzT2LIkCGqSxLFLcxGld6A0GAdDi9IVHUtoDy+SyjR6JtZx6FvMgGQfg7IXIUGmy02DQYjAnVaNskieS8kUGdTnJnYM0D6JbmI+DLyx8Q9XyrQWux6pecoqWHIn2upxJdi4xcqWePuz6uz8OXveCX48viVjk2RD2LOnDmCx8PCwtC6dWu8+uqr7LH3339fpagUiiW2ZNf2dNRuMbdlSzo3FQDDMNBo/XGuXRwGDU/E7bffprgd7iLIXCsjy0Ajeo7Uoi+3pZ/bDrnv5N4TRUksxkcoA7xYKgBuygFHWRnlgpjJ2BdsOWpxvlLFRkk6BKFz1KZREAuu55c7keqbZu6m+DqKlKVWrVoJHr/zzjsdKgzF91GyyHpCDiJHo3ZMas43GAzYunUr1q1bh9raWgDmz+Z3dT1wXq/DobIawevE7gV3EZw7+mas2JGPtFE3W1wrtSBzlRiS3JEsovw++S4hsmiTYG8AiO/RwSKWhrRDdpYB1xUR7jnceeT266jdjFzlZGBEmFX/3PxG6TnCZUek5lGJAqIkRxUffp9iP07IeRm5RVgxOlSyb1/8zFIoXBQpS4sWLXK2HJRmgtpfvRRpDh8+jFWrVgmmArhRokgqIH4vuIvg5EFd0DvwMuLiLDN4Sy3IpN1z1XrWPSVm9eC3w3UHkfNzT1awcT7cdrg7y+SeK7Ft8aQtNW4sobGStAncOnNcWduFBCD6hSwkxYZbFBmWmke+AiIkl5CSwlUQhy3eaTUOfp98hZIcI+elxEcCsEwEK9a3Urwx/QaleaN6K1B9fT0YhkFw8LWgzLIybN++Hd27d8cdd9zhcAEpvgU11zuGc+fOYc2aNdi7dy8As/V32rRpGD16NOt3l1vMBkaE4Vy13iorNHfxvOWNn2E0GDCvodQie7NU21LuKbGFmtsugRuLJBT0zF9slT5X3GvtdWORNi7XNkDfZGKD0fnKTPQLWWx9PK6yJHePhOKC1LjXhM4X65N/PvlnjumwVpbsgf5oongbqpWlJ554AqNHj8bkyZNx5coVPPDAA/D390dlZSXmz5+PKVOmOENOioej9Jei3K/e5obaAOhZQ7tAV7ofmzZtQlNTE/z8/JCUlITJkycr3uFGEMoKzV+cq6+5xTJyi0RLXaStP4SsI+XoE94Gl2obJZM/qrFGCClEYu2oaZf7DPLdhPx+AfOOwBU7LiCNpzDy25JKOJkUG46sI+VIig1XJCNgncRSTLkVu1ZobFK48ocM/dFE8TZUpw44duwYbr31VgBAdnY22rdvj//9739YsmQJ1q5d63AB1TBixAjExMRY/Fu6dKnkNQzD4L333sMdd9yB2NhYTJs2DX/99ZeLJPYd1OZOUXu+r6J0HtL/dwq1JXn4ZsV/8PXXX6OpqQn9+/fH6Bnz8emlaGw+eskhyf74QbvB/n7QALilW6ho+1lHymFkzJm7bbmnQqki1LppbBk7CWzm5i8S2in2n++Po6LOhIzcItG2kodcLykipACsmDwAhYuSLKxKSuTjuhoPllTCyAAHSyolryNKllBuJql5cmZuJXf2RaE4AtXKUn19PUJCQgAAu3fvxpgxY+Dn54e4uDiUl5c7XEC1pKWlYffu3ey/1NRUyfNXrVqFTz75BK+++io2bNiA9u3b49FHH0VNjXBQLEUYtblTPCXXiruzCSuZh8LCQsSc/h7dyn6GtrEGHTt2xAsvvIDXXnsN647VsQqKWgWUJOyM79GBnQOuPMlDItA2JAAMgD9Kq0TbT4oNhwaAv59GlSWDQJQWbv4hubHw75styvfAiDBoNbCw1PDbWZpdACMDaACkxEdJtudoBYDcC5JmQOlnhq9k8d+jP1IoFPWodsN169YNO3bswOjRo7F7925Mnz4dAHDp0iXVbgBnEBISojg5JsMw+Pzzz5GSkoIxY8YAAJYsWYKhQ4fihx9+wKRJk1T1bTQaVcvr6ZAxyY1t8qAubA0pJfOg9nxnkZ5zCmVV9UjPOSVYA0vp+Lms21eKjNwipMRHYcrgbpLnSs1DdXU1vvjiC+zYsQMMwyAwMBATJ07EuHHjEBAQAJPJhJT4yGt9md1D5P9q7kH82znsHOQ+n2Ahz+w7b8L7P5/E7Dtvgp/GT7D95Q/G4mDJZZRV1SMkUAvGZMKwxT8rGj8ApMRHYun2v6ABw7bNHZfQWPj3jXv+2j3Fiub/YMnla5aay2wf/H5JErqQAA0eurWzzc+q0meCnHdLt1D8UVqFlPgoTB7UBUajUfYzw70WYHBLt1Ck55wCYzKxfcrNqxC2fAZ8jeY+B748fqVjUp2Uctu2bZg7dy6MRiNuv/12rFmzBoC5wO6BAwfw8ccfq5fWQYwYMQKNjY1oampCp06dMHbsWMyYMQMBAQGC558+fRqjRo3Cpk2b0Lt3b/Z4amoqWrdujSVLlijqlyS1ongf2YV12JRfgwk9WyIxuoVD2kzJuoCKOhM6tPBDRlJH1dcbjUYcOHAAOTk5aGhoAGDOnD969Gi0bt3aZrnExip0XM28ZBfWYd3RqwADTOnXCpvya+wavz1jAZTPv5L5AKBoHqTkyS6sw6o/roAB0NJfg8/G3yDaDpHdTwOYGKiaQ/647X0OKZTmhEOSUnIZO3YsBg4ciIqKCvTs2ZM9fvvtt2PUqFG2SekgHn74YfTu3RutW7dGXl4eli1bhjNnzuDNN98UPL+iwhzY2q5dO4vj7du3t8ml2K9fP5/MbpqXl6d4bGqsKp5AXBwwT+J9teMHgLSG63PA33Ivx+HDh/HZZ59ZpAKYOXOmhTJvK2nbc1BRZ0JWUSPmTRzKHheaA3LuD4WNSIxuITn+tO05qGlk0Dk0CPMmDkXXfdLjd8QzInXflM6/UBvr9pVi9aFzMDLA6kNX8J97euPX8Z1lnwEyX1+d0COrqNFibGnbc1gLlQEaxMXFicpEZOdalsgYxOaNHB/SvaPFNfY8h1xs+Qz4Gs19Dnx5/GRscqivIgqgQ4cOVq6u2NhYW5qS5b333pPNCr5hwwb069ePdQkCQM+ePdG6dWukpaVh7ty5CAsT30HCLwFhYwUYaLVan3uQCErHlpFbjLKqemTkFovunuLjiJwrzs7boubeThsaqXjsBH4qgNatWyM5OdkiFYC9pCZ0Z+dIrk1yLsmxQ8YvNM/8duXGr/YZUXtvNX5+ADQ4UFKFjNxiVc9ERm4xm+jSyJj/JoqJ1DNA5sCcKdxybKkJ3fHK5qNgAATprrchNC6puRObN3IcqMav80eyx215DqXw5e83pTT3OWjO47dJWTpy5Ai2bduGs2fPoqmpyeI9R5c7mTp1Ku6++27Jc7p0sY41AcD+gistLRVUlojCd/HiRXTseN1MfenSJbRv395GiZs3tmwJdkTOFW/N26LX67FhwwZs3rzZ7lQAXLgLMXA9B5HQln6h2mD8HDtm60UxahsMVgki1SYnFHpGxBQiuS35Qtfxk2KqeSa4OaJyT1agtsGAp786jL2nhFMHEIQyhXPfI3IJZciWko9fDkbos0W34VMozkf1brisrCxMmTIFhYWF2L59OwwGA06dOoW9e/eKlkWxh7Zt2yI6OlryX2CgQOEigC0kKhbw3aVLF3To0AG//vore6yxsREHDhzAgAHKt/hSrmPLjiBH7IzzlN11SmEYBrt27cITTzyBb775hk0FsGLFCsyaNcsmRYm7Q4y7EMvtgBLajcYnI7fIorixPfMs9IyIySi1s0vsOvIsJMWGS8oqtBOSyLZi8gCEBOpQpW/Cj3nnZFMH8K8HYNG20JiVPLN8hUros6XmM+fu3Z8UireiWlnKyMjACy+8gI8++gj+/v546aWXsHXrVtx111248cYbnSGjIg4dOoRPP/0UJ06cwOnTp/Hjjz/i1VdfxYgRIxAefj0R3NixY7F9+3YAZvfbww8/jI8++gjbt2/HyZMn8cILLyAoKAj//Oc/3TWUZocjtlx7U96WwsJCzJ8/H8uWLcOlS5dwww034MUXX8Rrr72Gbt1sjy3h50giC7HQFnkuqQnRbC02sYU7JT6KTTPgaKQSKPK3zwvJTsZJFAEArMIjpLgQ5JRI0vbd/TqhQws/wdQBYsqHki36Sp5ZIYXKHoWHLxdVnigUZah2w50+fRrDhw8HAAQEBKCurg4ajQbTp0/HI488grS0NIcLqYSAgAD8+OOPeP/999HY2Ijw8HA8+OCDmDlzpsV5xcXFuHr1Kvv3rFmz0NDQgIULF6K6uhr9+/fHmjVrPCINAsW9mC00p5AUFQCJmFzFVFdXY+3atdi+fTubCuCBBx7A+PHjRXdsqkGssCmxzoglM1TiRpsyuBumDY1kS5AszS5wWIwYsWxxa8gplU2s5puY1UqqRhrfpWfpijwsGCSttG1bERq/PS5nvlze6r6mUFyNamWpTZs2bGXzG264AX/99RdiYmJw5coV6PV6hwuolD59+uDrr7+WPa+goMDib41Gg6eeegpPPfWUs0SjeCnmhaQem/IbJXfMyWEwGJCVlYUvv/yS/ewMHz4cjzzyiGRsnNrAZrFSMo6MaSFtXbhSjyp9E97MOi4rm9w4HCWf2pgeviJii+KgtG1HYs988eWi8U4UijIUK0svvPACXnrpJdx6663Ys2cPYmJicNddd+HNN9/E3r17sWfPHtx+++3OlJVCcSnmhcRsWbKVQ4cO4eOPP2ZTAURFRWH27NmKUgHY+quf72rhl++wxyJEFtvI+VkAgPomk93jcJRiIdaO0vZtURzsld2W++FIRcyZSh2F4ksoVpY2b96MuXPn4pVXXmET5T3++OPQ6XQ4ePAgRo8ejSeeeMJpglIoriZ5SAQmD+piU8LRc+fOYfXq1di3bx8AcyqAadOmYdSoUYq33tr6q597HVFUFmw5CsBxbpd7+isrDGtLQVd34Q7FQex+ODsVhhKIDCnxkegtvIeGQmk2KA7wJrmHQkNDccMN5gy0fn5+mDVrFhv03aZNG+dISfEIaDCoPHq9HmvXrsWTTz6Jffv2wc/PD/fccw8yMjKQmJioKkeJUACwknvAvS41IRpaDdgt9GI7sNTe2xWTB2DhuL44WFKJe9/bjegXspC2/pDVeULFapXiKc/bun2lSMm6gHX7SlVfm7b+kOjcAOI74rhKlCvmQagPIoOSXYAUiq+jajccP3kjpXlBi3CK44xUAEIouQfchS95SAQWjutrURxXaAeWLfeWXHOkrBpGBsg6Yp313taUDiS/ErGKkUVcSnFwllKRkVukOHUAn6wj5aJzA4jviOPOmys+d1IpGOQKCFMozQFVAd6JiYmyCtP+/fvtEojiudBgUGEKCwuxcuVKnDhxAoB548OMGTMwePBgVT8wlLhelNwDvuuN714SzsKt/t6mJkSzCSMByLrk1EB28AGwSCwp5Ubkv2eLK0vompT4KKzYkW+T0pAUq8xdyUco+NyZnzuh+89PTEqhNGdUKUtPPfWUUxJPUryD5hgMum5fKVbsEM7e7MhUAHKZqglK7gFXiRGKhRF6z5Z7y89MLXS9rTFS3EzaB0sq2UVcaru/rdviuW0IBcenxEciI6mjYOoAOYVsxeQBWDHZvgS3rvjcNcfPNoWiBlXKUlJSklXRWQrFl+G6YIiyZGsqACnkMlWrQay8BgAszS6AkQE0UNaPlLJI+pJLqmiLVUTpzjauciOUIVtJ3/xknkuzC1DbYMDS7AJU6ZuQkVuEFaNDJa/lWvGcjbODvz0huJxC8TQUxyzReCWKLyMW75ISH2WRvfnQoUP417/+hdWrV6O2thZRUVFYvHgxnnvuObvqCcplqpaTUy1tgv0VBY4LxeuokcHZmdWlYqKU9s1tI3lIBFvmBDCXdrmlW6hVgDeZA5IdnVjqpCDXpK0/ZNc9dHYME41NpFCsUb0bjkLxRcQWiCmDuyEjqSNGRATgzTffxIIFC3D69Gm0bt0aTz75JJYtW6YoZ5IcShd2pQsZ1+LBXZTnJsagc2gw5ibGKGqXqyySxX5pdoHixdTZO7kcWSoHAKsAkTn6df4I/FFaZaUwkvk6WFJpEUAvBbkm60i5XcqIs+sgeludRQrFFShWlvLz86kLjqIIT9nyrQZSH622wWAht16vx88//4ynnnrK7lQASpCbO6ULGT9lAIGvGJB+xGqQZeQWYULPlpgyuBu72APKi+l6k5WCqwBxFTC+dRGwtkapsWCJFfhV+rlxtrXO2e1TKN6I6nInFIoc3lhvir/Taurgbti1axc+/fRTXL5s3gnUv39/zJo1y65it3I4Ktu1VNwSvx/uedy2SczOuqMNmDfRuvacErxpB6VY8DgJ8D7eAIsyMrYExEtd442fGwqluUCVJYrD8aYFkguRe1LPAMyfP59NBRAaGoqUlBTcfvvtTo/dc+TcSS3OQlm+RRdpRr49W2TwNMSCx0mAd0ZuEcqq6hUpM7YESQ+MCMO5aj0GRoTZNQ4KheJ4qLJEcTjetEBy+WfPNqj6/Ti2f3I9FcD999+Pm266CbfeeqtLNjlw505owXXUTiUleXzmJsbA3tp4jsaVO7WIIpkSHwngMlLio7Bs+1+sq1aJlUjNLrmDJZUwMuZXCoXiWajK4E2h+CIGgwFbtmxBamoqfvrpJzAMg+HDhyM9PR33338/dDr3/KYQivdxRgwQP46J7NYCgNznE5AY3ULwOnfEprkyBorMy5TBZrfrlMHd2J1ycv2LxYzJXUMDqykUz4QqS5RmDT8VQHR0tGQqAFcqCEKLp9oFVWy7ulQtMKW7tVxdvwywT6EQklGt3Pz+xa7nl5lRAg2splA8F+qGozRLzp49izVr1mDfvn0AgNatW2PatGkYNWqU5A43VwbhCrkzlbo4ibuqtsGAKn0TzlXrJcuGZO4tQW2DAaHB/ojv0cEia7YYquKeHIQaF2/m3hIszS4AQFyK1jKqlVsqKaat94pCoXg+VFmiNCv0ej2++eYbbN68GQaDAX5+fvjnP/+JSZMmKSp266zgdUfE4pA2BkaEsQVcQ4P90Tk0WLRsyMCIMAxbvJNVqjqHBluU5zAajaL9ubp+mRq4ZV0AWCh1fEudPXJ762YGCoWiDqosUZoFDMNYpQKIi4vDzJkzVaUCcJa1QMxCIaVE8d8jbRArklZjtqhIXUeuIUqVkkVfSCZPs6KQ8jEamLOVi233t1duJdfT8iEUivdDlSWKz3Pq1CmsXLkS+fn5AIBOnTrhsccew+DBg91exodrDQKs67VJuXn47wkVn5UrbmtL7iR35gNSqnjYMi5nQfMnUSjeD1WWKD5LVVUVMjMzsX379VQADzzwAMaPH4+AAPdvh+e7isiONC5SuXf4LiClVhK+IqF2AXen60mp4uFuSxdXqXPUfFELFYXiPqiyRPE5DAYDsrKy8OWXX6K2thYAMHz4cEyfPt2jSvYQV5FWY21RIvBz7/AXTFsWTVe4npxFakI0lmYXKMp1JIYrlA6uUueoHW7UQkWhuA+aOoDiUxw6dAhpaWmCqQA8SVECrm9DXziur2SmbW4skVyeIbmt8Pa+726Sh0QoznUkhityNTkjZxLNw0ShuA+qLFF8grNnz+KNN97AggULcObMGbRp0wZz5szB0qVL0bt3b3eLJwg3r45Uvh6uZUJuwZRTBOx93xOQmwNHFSO2B7GCxY5ok1qVKBTXQ5Ulilej1+vx+eef48knn8T+/fvh5+eHe++9F+np6RgzZoxkziR3IZUQUk5JkVsw5RQB8j5JGcBfxJ2lSDjSYiU3B3Jz6UqlwxuUTwqFIg9VliheCcMw+N///ofU1FRs2LABBoMBcXFxWLFiBWbOnKkoZ5K7EFpAHaWkyCkC5P2DJZWCi7itioScMiQ0Zme5/DzJXeVJslAoFNuhAd4Ur+Ovv/7CqlWrPCIVgC3BwkK7o1wdNC22Q0vpeMRyPIkFHwv156yAZXfvhOPiSbJQKBTboZYlitdQVVWF9957D3PnzkV+fj4CAwORnJyM999/H0OGDHFLziRb3CyeEHsiJoPS8fDPk7OgCPUn5xK0B08PVFeKr4yDQvF2qLJE8XgMBgO2bNmClJQUNmfS8OHDkZGRgQcffNCtOZPc4WZx5gKqdDz882xRAMVcgo4Yn6/ECvnKOCgUb4e64SgezaFDh7Bq1SqcOXMGABAdHY1Zs2Z5zA43ITeLmhIlSlDr8rIHpW4jR7qX+C46R4zPV2q2+co4KBRvhypLFI/k7NmzWL16Nfbv3w8AaNOmDaZNm4aRI0d65A43LmpKlKhtDwBqGwwIvVbvTI7MvSVYml0AQLhOnDvhKoHc7OWOUBB8JVbIV8ZBoXg71A1H8Sj4qQC0Wq3HpwLgI+XKssVtx70mPacQVfomhATqFC2i5Hy5JI7uiI0RczF5QkwXhUKhcKGWJYpHwDAMcnJy8Nlnn+Hy5csAgAEDBmDmzJno2rWrm6VTh5Q1wBZLAf8aNVYXUh6kwWC0KBHiSteelGzUxUShULwBqixR3I4npQLwFMRim9QqW+T8YYt3WihDfOXIHYoLdTFRKBRvgSpLFLdRVVWFtWvXYseOHWAYBkFBQXjggQcwbtw4t+5wcxVSwd5ylh61+ZAGRoQBuF6wl68cUcWFQqFQxKHKEsXlGAwGZGVlYf369airqwMADB8+HNOnT/e4YrfOREohkrP0KHWbkfMAWARRU+WIQqFQlOMzytK+ffvw8MMPC773zTffIDY2VvC9+fPnY9OmTRbH+vfvj6+//trhMlKAP/74Ax9//DGbCqB79+6YNWsWevXq5WbJzKStP4SsI+VIig3HiskDnNqXlEIkp8wodZu5wr1mSzoEd8C1sh0sqfR4eSkUiufgM8rSgAEDsHv3botj7777Lvbs2YN+/fpJXnvnnXdi0aJF7N/+/v5OkbE5U15ejjVr1nh8KoCsI+UwMuZXZytLaqw7fIXEHfmQxHBHcLgtEDnPVethZOBSeb1FoaRQKML4TOqAgIAAdOjQgf0XGhqKnTt3YuLEibJBwkLXUhyDXq/HZ599hjlz5rCpAMaNG+e0VAD2boFPig2HVmN+dTfcsXhyJmel6RDcXbqDyJkUG+7yrOuefP8oFIo8PmNZ4rNz505UVlbivvvukz13//79uP3229G6dWsMGjQIzzzzjE2xM0aj0RZRPRoyJrVjYxgGu3btwueff47KykoAQFxcHB577DE2FYAz5is95xTKquqRnnMKkwd1UX398gdjsfxBs8vWaDTaPH451u0rRUZuEVLiozBlcDfBc7hjSYmPunZ+pEufMyXjnzyoCzvXUufZe2/shTGZADAYFBFqcY/l7oUjnoGU+EjZ+6fkmXAHzvoMeBPNfQ58efxKx6RhGIZxsixuYdasWQCAVatWSZ73448/okWLFggPD8eZM2fw7rvvwmg0YuPGjYp3ZBmNRhw+fNhekX2G8vJybN26lY1LCgsLQ2JiInr06OH0VADZhXXYlF+DCT1bIjG6hVP7UoqQTClZF1BRZ0KHFn7ISOqo+Dq1/XgS2YV1WHf0KsAAU/q1crqM/PkQm3Ml98IVeIocFEpzJC4uTtLT4fHK0nvvvYf3339f8pwNGzZYxCWdO3cO//jHP/DOO+8gMTFRVX8XLlzAiBEj8N///hdjxoxRdA1Rlvr16+dR8TeOwGg0Ii8vT9HYqqqqkJmZiZ07d7KpAO6//37cc889XpsKQM34xYh/OwdlVfXoHBqE3OcTADjHiiDUj704YvxcnCGj0r7E5lyJZcmRcyCGJ1uWXDF+T6a5z4Evj5+MTU5Z8ng33NSpU3H33XdLntOli6VJ/9tvv0VoaChGjBghcoU4HTt2RHh4OP7++2/V12q1Wp97kAhSY2tqakJWVha+/PJLNhVAQkICHnnkEZ9JBWDPvU1N6M4G95I2pg2NxLShkY4UUbAfR+GoZ9uZMsr1JTbnSu+Fsz/fzngmHIkvf78ppbnPQXMev8crS23btkXbtm0Vn88wDDZu3Ijx48fbtKutsrISZ8+eRceO1AyuhIMHD+Ljjz9GWVkZAM9LBeAJ2LojTe0OKkfufCN9p8RHonegQ5oE4Nr8TjSXFIVCcRQeryypZe/evThz5gzuv/9+wffHjh2L5557DqNHj0ZtbS3ef/99jBkzBh06dEBZWRmWL1+OsLAwjBo1ysWSexdiqQBGjRoFPz+f2WTpVty5JZ/0nZFbhBWjQ13aN4VCoXgaPqcsbdiwAQMGDEB0tPC24OLiYly9ehWA2aR48uRJbN68GVevXkWHDh0wePBgLF++HC1btnSl2F5DXV0dvvnmG2zZsgUGgwFarRb//Oc/8dBDD3ntnHlqDhx3FpolfafERwK47PL+KRQKxZPwOWVp2bJlku8XFBSw/w8KCsLq1audLZJPYDKZkJubi88++wyXL5sXzwEDBmDmzJlsKgBvRciC4yw3lBrc6UYifZs3L1BliUKhNG98TlmiOJ6ysjKsX78eJ0+eBAB06tQJM2bMwG233eb0VACuQMiCQ91QFAqFQiFQZYkiSmVlJT7//HP8/PPPAMyWuAceeADjxo3z2lQAQghZcKgbikKhUCgEqixRrGhqasIPP/yAr776ik0FMHz4cEyfPt1nUgHI4UluKE+NqaJQKJTmAt22RLHg4MGDSEtLwyeffIK6ujpER0fjsccew9NPP+1SRUmqjpi7a4y5GlpXzDdpbs8xheLNUGWJAsCcCuD111/HwoULUVZWhjZt2mDOnDl466233BLALaUgNDflQWmhWop30dyeYwrFm6FuuGaOXCoAdxVOlNo2784t9e5Aza446rLzHprbc0yheDNUWWqmmEwm5OTk4PPPP/fIVABSCoK9W+p9WaFwZyJLT8YT7znNME6heA9UWWqG/PXXX1i5ciWbc6pTp06YOXMmBg0a5BOpAOTwZYWiuVgruMoPAFlFSO6ee6IyRaFQPAeqLDUjKisrsXbtWuzYsQOAORXAgw8+iHHjxtlUR89b8WWForlYK/jxPnLKr9w992UFmkKh2A9VlpoBQqkAEhIS8MgjjzSbVABcmotC4cvwlR855VfunvuyAk2hUOyHKks+zsGDB/Hxxx+jrKwMANC9e3fMnj0bPXv2dLNkFIrt8JUfKUVIiYuNKtAUCkUKqiz5KOXl5Vi9ejUOHDgAAGjTpg0efvhhjBw5En5+NGMEpflAXWwUCsVeqLLkY4ilApg0aRJCQkLcLR6F4nKkXGw0sJtCoSiBKks+glgqgFmzZqFLly5ulo5CcR9SLjZqdaJQKEqgypIP0NxTAVAotkKsTgMjwjBs8U5qYaJQKIJQZcmLqaysxOeff46ff/4ZgDkVwEMPPYR77723WaUCoFBshVidhi3eSS1MFApFFBrp64U0NTVh06ZNSE1NZRWlf/zjH0hPT8fEiRN9WlGixUcpfOSeCSXPDK2/R6FQpKCWJS+juacCoDEmFD5yz4SSZ4amDqBQKFJQZclLoKkAzNDkgZ6Dp+wkk3sm6DNDoVDshSpLHk5dXR2+/vprfPfdd2wqgHvuuQcPPfRQs0wFQC0AnoOnWPnkngklz4ynKH4UCsUzaT4mCS+ksLAQqamp2LhxIwwGA2655Ra89957eOyxx5qlokTxLLw5zocfx8SvNUehUChcqGXJgzl48CAqKyvRqVMnzJo16//bu/e4mPP9D+CvUYqUSxfRBbvRoFLJbkrFcds5VsfKZctt0brveoStTUfYQiGPbFrklrKxdUS7i5Pl7HqsilwPUWtdWpdKUTglTGp+f/g1a1aNiplvTa/n4+Hx0Gfm+533+/v1MK8+n+93Bv369dOojwLgb/NNW1Oe5fvrrBiX6ohIGYalRszLywu9e/eGWCzWyDvcGssyDmm+vwbzv4ajphz8iEj1uAzXiGlra8PW1lYjgxLQdJZxDl8vh8faY0324wr4cQuvLrNN6t8V6YGDGZCIqE4YlkitXn7jbipvWPt/K0Pew6dN9noWXo/TdII5ETVODEukVk3xjXt0T32Yt2+l8Eb7prM16pztYVDgTBIRvRmGJVKrpvjG/YGVHn71H6TwRvumoU+dobG5BAUuNxKRqjAskVppyhv3m4a+phgaG7umOGtJRE0D74ajeuMt/29+9xTvvnr7ePs/EakKwxLVG2/5p8aIAZSIVIXLcFRvXEIiIqLmhDNLVG/8DZ6IiJoTziwRERERKcGwRERERKQEwxIRERGREgxLREREREo0mbC0adMmeHt7w97eHv369avxOfn5+Zg9ezYcHBzg7OyMFStWQCqVKt2vVCpFaGgonJ2d4eDggNmzZ+Pu3buqaIEaiJ/MTEREQmoyYamiogISiQQ+Pj41Pl5ZWYlZs2ahvLwcu3fvRmRkJA4fPozVq1cr3e/KlStx5MgRREZGYvfu3SgvL8esWbNQWVmpijaoAfjJzEREJKQmE5bmz5+PqVOnwtrausbH09LScO3aNaxduxa9e/eGq6srAgMDkZSUhLKyshq3KS0tRXJyMgIDA+Hq6orevXtj7dq1+P3335GRkaHKdqge+LlOREQkJI35nKX//ve/6NGjB0xNTeVjbm5ukEqluHTpEvr37//KNpcuXUJFRQUGDBggHzM1NUWPHj1w/vx5uLu716sGTZyNqu6pIb3tzryFzb/ewGyPdzHBuUuDa/B5zwI+71k0uI438Sb9a4Lm3j/AY9Dc+wd4DDS5/7r2pDFh6f79+zA2NlYYa9euHVq2bIn79+/Xuk3Lli3Rrl07hXFjY+Nat1EmKyur3ts0FQ3pLepoEe6VVyHq6G/orVuigqrUR5PPbV009/4BHoPm3j/AY9Cc+xc0LG3YsAHR0dFKn7N3717Y2dnVaX8ikahe47WRyWT1en41Ozs7aGlpNWjbxqqyshJZWVkN6m3+sz9nlhwcGj6zJKQ36V8TNPf+AR6D5t4/wGOgyf1X9/Y6goaliRMnYsSIEUqfY2FhUad9GRsb48KFCwpjjx49QkVFBYyMjGrdpqKiAo8ePVKYXSouLoajo2OdXvdlWlpaGvcPqVpDepvs+g4mu76joorUS5PPbV009/4BHoPm3j/AY9Cc+xc0LBkaGsLQ0PCt7MvBwQGbN29GUVEROnbsCABIT0+Hjo4ObG1ta9zG1tYWLVu2RHp6ujy0FRUV4erVq/D3938rdREREVHT1mTuhsvPz0dOTg7y8/NRWVmJnJwc5OTk4PHjxwBeXMzdvXt3BAQEIDs7GydOnMDq1asxfvx46OvrAwAKCwshkUhw8eJFAICBgQHGjBmD1atX48SJE8jOzoa/vz+sra3h6uoqWK9ERETUeDSZC7yjoqKwf/9++c8fffQRACA+Ph7Ozs7Q0tJCTEwMvvrqK/j4+KBVq1YYOXIkvvzyS/k2FRUVyM3NxZMnT+RjQUFB0NbWhp+fH54+fQoXFxeEh4c326nGxuTbkzex6dh1zBlkhUn9uwpdDhERNVNNJiyFh4cjPDxc6XPMzMwQExNT6+MWFha4cuWKwpiuri6Cg4MRHBz8Vuqkt+flD6NkWCIiIqE0mWU4an74YZRERNQYNJmZJWp+JvXvyhklIiISHGeWiIiIiJRgWCIiIiJSgmGJiIiISAmGJSIiIiIlGJaIiIiIlGBYIiIiIlKCYYmIiIhICYYlIiIiIiUYloiIiIiUYFgiIiIiUoJhiYiIiEgJhiUiIiIiJRiWiIiIiJTQFroATSCTyQAAlZWVAlfy9lX3pIm91QX7b979AzwGzb1/gMdAk/uv7qn6fbw2ItnrnkGvJZVKkZWVJXQZRERE1AB2dnbQ0dGp9XGGpbegqqoKz58/R4sWLSASiYQuh4iIiOpAJpOhqqoK2traaNGi9iuTGJaIiIiIlOAF3kRERERKMCwRERERKcGwRERERKQEwxIRERGREgxLREREREowLBEREREpwbBEREREpATDEhEREZESDEtUZ7m5uZgzZw6cnZ3Rt29feHt74+TJk0KXpVbHjh3DuHHj0KdPHzg7O+Ozzz4TuiRBSKVSjBo1CmKxGDk5OUKXoxZ37txBUFAQBg8ejD59+mDo0KGIioqCVCoVujSVSkhIwODBg2FnZwcvLy+cOXNG6JLUIiYmBmPGjIGjoyNcXFwwd+5c3LhxQ+iyBBMTEwOxWIyVK1cKXYogGJaozmbNmoXKykrExcVh37596NWrF2bPno179+4JXZpaHD58GAEBAfDy8sL333+PPXv2YOTIkUKXJYg1a9agY8eOQpehVjdu3IBMJkNISAgOHjyIxYsX47vvvkNkZKTQpanMoUOHEBYWhjlz5iAlJQVOTk6YMWMG8vPzhS5N5U6dOoWJEyciKSkJsbGxqKyshK+vL8rLy4UuTe0uXryIxMREiMVioUsRjoyoDoqLi2XW1tay06dPy8dKS0tl1tbWsoyMDAErU4+KigqZu7u7LCkpSehSBHfs2DGZRCKRXb16VWZtbS3Lzs4WuiTBbN26VTZ48GChy1CZsWPHypYuXaowJpFIZBEREQJVJJzq/wNPnToldClqVVZWJhs+fLgsPT1dNmnSJNmKFSuELkkQnFmiOunQoQOsrKyQkpKC8vJyPH/+HImJiTA2NoaNjY3Q5alcdnY2CgsL0aJFC3z00Udwc3PDp59+iqtXrwpdmlrdv38fwcHBWLNmDVq1aiV0OYIrLS1Fu3bthC5DJaRSKS5fvgw3NzeF8QEDBuD8+fMCVSWc0tJSANDY812bkJAQDBw4EK6urkKXIiiGJaoTkUiE2NhYZGdno2/fvujTpw927tyJbdu2oW3btkKXp3K3b98GAERHR2POnDnYvHkz2rVrh0mTJuHhw4fCFqcmMpkMgYGB8Pb2hp2dndDlCO7WrVv49ttv4ePjI3QpKvHgwQNUVlbCyMhIYdzY2LjZLL1Xk8lkCAsLg5OTE6ytrYUuR20OHjyI7OxsLFq0SOhSBKctdAEkrA0bNiA6Olrpc/bu3QtbW1ssX74cRkZGSEhIQKtWrfCvf/0Ls2bNwt69e5vs9St17b+qqgoAMHv2bHzwwQcAgLCwMHh4eCA1NRXe3t4qr1VV6noMzp8/j7KyMsyaNUtNlalHXft/OSAWFhbi008/hUQiwbhx41RdoqBEIpHCzzKZ7JUxTRcSEoLff/8du3fvFroUtSkoKMDKlSuxY8cO6OrqCl2O4BiWmrmJEydixIgRSp9jYWGBkydP4tixYzh9+jT09fUBADY2NsjIyEBKSgpmzpypjnLfurr2//jxYwCAlZWVfFxHRweWlpYoKChQaY2qVtdjsGnTJly4cOGVWaUxY8bA09MTq1evVmWZKlPX/qsVFhZiypQpcHBwQGhoqKrLE0yHDh2gpaWF+/fvK4wXFxfD2NhYoKrULzQ0FD///DO+/fZbdOrUSehy1Oby5csoLi6Gl5eXfKyyshKnT59GQkICsrKyoKWlJWCF6sWw1MwZGhrC0NDwtc978uQJgFd/yxSJRPJZl6aorv3b2tpCR0cHubm56NevHwCgoqICeXl5MDMzU3WZKlXXY7BkyRL4+fnJfy4qKoKvry8iIyNhb2+vwgpVq679A38GJRsbG4SFhaFFC829kkFHRwc2NjZIT0/HsGHD5OMZGRkYMmSIgJWph0wmQ2hoKI4cOYJdu3bB0tJS6JLUqn///vjxxx8VxhYvXox3330XM2bMaFZBCWBYojpycHBA27ZtERgYiHnz5kFXVxdJSUnIy8vDoEGDhC5P5fT19eHt7Y0NGzagc+fOMDMzw/bt2wEAEolE4OrU46+hUE9PDwDQpUuXZvEbd2FhISZPnozOnTvjyy+/RElJifwxExMTAStTnWnTpiEgIAC2trZwdHREYmIiCgoKmvSyc1199dVXOHDgADZu3Ig2bdrIr9MyMDBoFjc36Ovrv3J9lp6eHtq3b9+srtuqxrBEdWJoaIht27Zh/fr1+OSTT1BRUYEePXrgm2++Qc+ePYUuTy0CAgKgra2NgIAAPH36FPb29oiLi2t2d8c0V+np6bh58yZu3rwJDw8PhceuXLkiUFWqNWLECDx48AAbN25EUVERrK2tsWXLFpibmwtdmsrt2bMHADB58mSF8bCwMIWlKWoeRDKZTCZ0EURERESNleYuuBMRERG9BQxLREREREowLBEREREpwbBEREREpATDEhEREZESDEtERERESjAsERERESnBsERERESkBMMSEdUqMDAQc+fOFboMBZmZmRCLxfjf//4ndCk1Onv2LDw9PWFjY9Pojt3bJBaLcfToUaHLIFILft0JURMnFouVPj569GiEh4erqZoX9XzzzTcYOnSoSvbv6OiItLQ0GBgYAAD27duHVatW4cyZMw3eZ2ZmJqZMmSL/WVdXF5aWlpgyZQo+/vjjeu0rPDwcPXv2xNatW+Xfn6cuL/chEonQpk0bWFpawtXVFVOnTkXHjh3f2mulpaXxq36o2WBYImri0tLS5H8/dOgQoqKikJqaKh/TtC/91NHRUdkX16ampkJfXx9Pnz7FL7/8guXLl6NLly5wcXGp8z5u3boFb2/vN/pyYalUCh0dnQZvX91HWVkZsrOzsW3bNiQnJyM+Pv614bquNPXLg4lqwmU4oibOxMRE/sfAwAAikUj+s7a2NpYtWwYPDw/Y29vD09MTBw4cUNg+NTUVnp6e6NOnD5ydnTF16lSUl5fX+FqXLl2Ci4sLNm3a1KBaq6qqEB0dDQ8PD9ja2mLUqFH49ddfFZ5z7tw5jBo1CnZ2dvDy8sLRo0chFouRk5MDQHEZLjMzE4sXL0ZpaSnEYjHEYjE2bNgAAEhISMDw4cNhZ2cHV1dXzJ8//7X1GRkZwcTERD6rZG5ujsuXL8sfl8lk2Lp1K4YMGYI+ffrgH//4hzyY3rlzB2KxGA8fPkRQUBDEYjH27dsHADh16hTGjh0LW1tbuLm5ISIiAs+fP5fvd/LkyQgJCUFYWBicnZ0xffp0AMC1a9cwY8YMODo6wtXVFf7+/igpKalzH++88w4+/PBD7NmzBx06dMDy5csVnpecnIy///3vsLOzg0QiQUJCgvwxqVSKkJAQuLm5wc7ODoMHD0ZMTIz88b8uw9X1vJ04cQJeXl6wt7eHt7c3bty48dp+iITGmSUiDSaVSmFjY4MZM2ZAX18fx44dQ0BAACwtLWFvb4+ioiIsWrQI/v7+GDp0KB4/fowzZ86gpu/XzszMxLx587Bw4UJMmDChQfXEx8cjNjYWISEh6NWrF5KTkzF37lwcOHAA3bp1Q1lZGebMmQMPDw+sW7cOeXl5WLVqVa37c3R0RFBQkMJsmp6eHrKysrBy5UqsWbMGjo6OePToUb2W6WQyGY4fP467d+/C3t5ePr5+/Xr89NNPWL58Obp164bTp0/D398fhoaGcHJyQlpaGiQSCebPn48RI0bAwMAAhYWFmDlzJkaPHo3Vq1cjNzcXS5Ysga6uLj7//HP5vvfv3w8fHx/s2bMHMpkMRUVFmDRpEsaPH4/AwEA8e/YMERER8PPzQ3x8fL2Oe6tWreDt7Y2wsDAUFxfDyMgISUlJiIqKwtKlS9GrVy/k5OQgODgYenp6GD16NHbt2oWff/4Z69evR+fOnVFQUIC7d+/WuP/6nLfIyEgEBgbC0NAQy5YtQ1BQEL777rt69UOkbgxLRBrM1NQUvr6+8p8nT56M48ePIzU1Ffb29rh37x6eP3+OYcOGwdzcHEDN10AdPXoUAQEBCAkJwciRIxtcz/bt2zFjxgx8+OGHAAB/f39kZmYiLi4Oy5Ytw48//ggAWLFiBXR1ddG9e3cUFRVhyZIlNe5PR0dHYTatWkFBAVq3bo1BgwZBX18f5ubm6N2792vrGzhwIIAXIbOqqgrz58/He++9BwAoLy9HbGws4uLi4OjoCACwtLTE2bNnkZiYiPfffx8mJiYQiUQwMDCQ17Np0yZ06tQJS5cuhUgkgpWVFQoLCxEREYF58+ahRYsXE/xdu3ZFQECAvJavv/4aNjY2WLhwoXxs1apVGDhwIHJzc/HOO+/U7aD/v3fffRcAkJeXByMjI2zcuBGBgYEYPny4vJdr164hMTERo0ePRkFBAbp27QonJyeIRCL5v4+a1Oe8LViwAO+//z4AYObMmZg5cyaePXsGXV3devVDpE4MS0QarLKyElu2bMGhQ4dQVFQEqVQKqVSK1q1bAwB69uwJFxcXeHp6ws3NDW5ubvjggw8ULty9ePEijh07hq+//hrDhg1rcC1lZWUoKipC3759Fcb79u2L3377DQCQm5sLsVis8MZpZ2dX79dydXWFmZkZhg4dCnd3d7i7u2PYsGHyvmuTkJCANm3aQCqV4uLFiwgNDUW7du0wYcIEXLt2Dc+ePZMvkVWrqKhAr169at3n9evX4ejoCJFIJB9zcnJCeXk57t69CzMzMwCAra2twnaXL19GZmamPJi97NatW/UOSy/PFpaUlKCgoAD//Oc/ERwcLB9//vy5/ML50aNHY/r06ZBIJHB3d8egQYPg5uZW477rc95eDuPVgbK4uFh+HIgaI4YlIg22Y8cO7Ny5U34NTevWrbFq1SpUVFQAALS0tBAbG4tz584hPT0du3btQmRkJJKSkmBpaQngxYxD+/btkZycjIEDB77RhccAFEID8OJNvHrs5b+/CX19fezfvx+nTp1CWloaoqKiEB0djb1796Jt27a1bmdhYSF/vEePHrhw4QI2b96MCRMmyMNGTEwMTE1NFbZTdkxqWtKsHnu5178GuaqqKvztb3/DF1988cr2Dbm4uvraIAsLC1RVVQEAQkNDFZYZAchnumxsbPCf//wHv/76KzIyMuDn5wdXV1dERUXV2E9dz5u29p9vO9XbVNdD1FjxAm8iDXb27FkMGTIEo0aNQs+ePWFpaYk//vhD4TkikQhOTk6YP38+UlJS0LJlS4ULdzt06IC4uDjcunULCxYskAet+tLX10fHjh1x9uxZhfHz58/DysoKwIuloitXrkAqlcofz8rKUrrfli1borKy8pVxbW1tuLq6IiAgAD/88APy8vJw8uTJetWspaWFZ8+eAQCsrKygo6OD/Px8dO3aVeFP586da91H9+7dcf78eYXQdO7cObRp0+aV0PUyGxsbXL16Febm5q+8Xn0/kuDp06dITEzEe++9B0NDQxgbG8PU1BS3b99+Zd/VIRl4cc5GjBiBFStWIDIyEocPH8bDhw9f2X9DzhtRU8KwRKTBunTpgoyMDJw7dw7Xr1/H0qVLcf/+ffnj1TMnWVlZyM/Px08//YSSkhL59S3VjIyMEBcXhxs3bmDRokUKd3LV5M6dO8jJyVH48/jxY/j6+mLr1q04dOgQbty4gYiICPz222/yzwby9PSETCZDcHAwrl+/juPHj2PHjh0AXp2RqmZubo7y8nKcOHECJSUlePLkCX755RfEx8cjJycHeXl5SElJQVVV1WuXroqLi3Hv3j3k5eXh3//+N77//nsMHjwYwIvgMH36dISFhWH//v24desWsrOzkZCQgP3799e6zwkTJuDu3bsIDQ3F9evXcfToUWzYsAHTpk2Tz+LUtt2jR4+wcOFCXLx4Ebdv30ZaWhoWL15cYzisqY8//vgDBw8ehI+PDx48eKBwN9znn3+OLVu2IC4uDrm5ubhy5QqSk5MRGxsLANi5cycOHjyI69evIzc3F6mpqTAxMalxZq4h542oKeEyHJEGmzt3Lu7cuQNfX1+0bt0a48ePx9ChQ1FaWgrgRQA4ffo04uLiUFZWBjMzMwQGBsovdH6ZiYkJ4uLiMHnyZHzxxRdYt24dtLS0anzdsLCwV8bi4+MxZcoUlJWVITw8HCUlJbCyssLGjRvRrVs3eT2bNm3C8uXLMWrUKFhbW2PevHlYtGhRrUtdffv2hbe3N/z8/PDw4UN89tlncHFxwZEjRxAdHY1nz56ha9euWLduHXr06KH0eEkkEgAvZqU6deqEjz/+WOGONT8/PxgZGSEmJgZ37tyBgYEBevfujdmzZ9e6T1NTU2zZsgVr1qxBUlIS2rdvj7Fjx2LOnDlKazE1NcWePXsQEREBX19fSKVSmJmZwd3dXWnIqu5DJBJBT08PlpaWGDBgAKZNm6awfDdu3Di0atUK27dvx9q1a6Gnpwdra2t88sknAF7cVbh161bcvHkTLVq0gJ2dHbZs2VLjazfkvBE1JSJZTQvqRESNxA8//ICgoCCcOXNG4z5gU5PxvJEm4cwSETUqKSkpsLCwgKmpKa5cuYKIiAhIJBK+4TZyPG+kyRiWiKhRuXfvHqKionDv3j2YmJhAIpFgwYIFQpdFr8HzRpqMy3BERERESvBuOCIiIiIlGJaIiIiIlGBYIiIiIlKCYYmIiIhICYYlIiIiIiUYloiIiIiUYFgiIiIiUoJhiYiIiEiJ/wNFiFy92mCIpwAAAABJRU5ErkJggg==",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"y0 = predict(model, X).numpy(force=True)[0]\n",
"y1 = predict(model, X_hat).numpy(force=True)[0]\n",
"\n",
"plt.plot([-9, 5], [-9, 5], c='0.3')\n",
"plt.scatter(y0, y1, s=1)\n",
"plt.scatter(y0[idxs], y1[idxs], s=2, color='r', label='AP-1 Targets')\n",
"plt.legend()\n",
"plt.xlabel(\"Task Logits Before Design\")\n",
"plt.ylabel(\"Task Logits After Design\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"id": "b4a5264b-3da6-4f0d-aee6-25389871ca02",
"metadata": {},
"source": [
"Looks like we are starting to get there, but are not as close as implanting ten entire motifs.\n",
"\n",
"This method has strengths and weaknesses. The weaknesses are that it may take many more iterations to get good results and that it gives potentially too much flexiblility of the model, allowing it to overfit to spurious correlations (e.g., finding a random mutation that is not part of the model but happens to make the model predict closer to the desired output). The strengths are that it does not rely on a set of pre-identified set of motifs and so can make changes that fall outside the set of motifs, e.g., adding in a lower affinity version of a motif or an alternate version of the motif that is actually stronger than the one provided. "
]
},
{
"cell_type": "markdown",
"id": "a369065f-723f-4021-b0cc-45a2035c1ddc",
"metadata": {},
"source": [
"#### Balancing Losses\n",
"\n",
"Is it possible to design a sequence that keeps the other tasks as close to their original values as possible while still increasing the predictions for AP-1 binding? To do this, rather than using a mask to ignore the outputs from non-AP-1 tasks, we will set the target values for that task to the original predictions from the model."
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "f5e43a8f-137e-4d19-9f57-a3adda8f30d8",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 0 -- Loss: 1.13, Improvement: N/A, Idx: N/A, Time (s): 0s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.95it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 1 -- Loss: 0.6533, Improvement: 0.4764, Motif Idx: 15, Pos Idx: 896, Time (s): 8.819\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.89it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 2 -- Loss: 0.5997, Improvement: 0.05358, Motif Idx: 2, Pos Idx: 905, Time (s): 9.001\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.99it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 3 -- Loss: 0.5819, Improvement: 0.01777, Motif Idx: 14, Pos Idx: 1063, Time (s): 8.694\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.96it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 4 -- Loss: 0.5713, Improvement: 0.01067, Motif Idx: 13, Pos Idx: 852, Time (s): 8.774\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.99it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 5 -- Loss: 0.5646, Improvement: 0.006629, Motif Idx: 12, Pos Idx: 704, Time (s): 8.688\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.92it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 6 -- Loss: 0.5586, Improvement: 0.006042, Motif Idx: 23, Pos Idx: 1550, Time (s): 8.893\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:09<00:00, 2.89it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 7 -- Loss: 0.5533, Improvement: 0.005298, Motif Idx: 3, Pos Idx: 805, Time (s): 9.019\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.97it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 8 -- Loss: 0.5494, Improvement: 0.003871, Motif Idx: 12, Pos Idx: 1193, Time (s): 8.761\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.96it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 9 -- Loss: 0.5464, Improvement: 0.003075, Motif Idx: 5, Pos Idx: 1588, Time (s): 8.781\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.97it/s]"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 10 -- Loss: 0.5447, Improvement: 0.001643, Motif Idx: 11, Pos Idx: 1803, Time (s): 8.754\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"y = predict(model, X)\n",
"y[:, idxs] = 3\n",
"\n",
"X_hat2 = greedy_substitution(model, X, motifs, y, max_iter=10, verbose=True)"
]
},
{
"cell_type": "markdown",
"id": "326e70ee-bf46-4b9b-9d01-33430e2e6ff8",
"metadata": {},
"source": [
"Unsurprisingly, we can see that the magnitude of the loss goes down because we are now averaging over two orders of magnitude more tasks. Interestingly, the model choose to include motif index 4 rather than motif index 2 in two of the iterations. This motif is the one for FOS-JUN on JASPAR but is considered to be a weaker AP-1 site by the Beluga model than the one we've been using thus far. Additionally, it looks like the motifs are placed in the sequence further apart than they were in the other example."
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "92c7a9ca-7d23-47d2-b942-e2423f6f89fc",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"y0 = predict(model, X).numpy(force=True)[0]\n",
"y1 = predict(model, X_hat2).numpy(force=True)[0]\n",
"\n",
"plt.plot([-9, 5], [-9, 5], c='0.3')\n",
"plt.scatter(y0, y1, s=1)\n",
"plt.scatter(y0[idxs], y1[idxs], s=2, color='r', label='AP-1 Targets')\n",
"plt.legend()\n",
"plt.xlabel(\"Task Logits Before Design\")\n",
"plt.ylabel(\"Task Logits After Design\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"id": "201fd87f-b7d1-4200-8267-c57c357a0fa1",
"metadata": {},
"source": [
"And it seems like we've reached a new compromise! The AP-1 related tasks are nicely separated from the other tasks (except for two of them) and, although they do not reach the desired objective value of `3`, the flip side is that the other tasks seem to be much closer to their original values.\n",
"\n",
"\n",
"#### Input Masks\n",
"\n",
"Sometimes, we may not want to consider substitutions at every position in the sequence. Perhaps we know that the most important part of the sequence is in the middle, or have known motifs in our sequence that we want to protect, or maybe we just do not need to be as precise as considering any possible possible. Regardless of the reason for having a mask, using one can speed up design significantly.\n",
"\n",
"Normally, the input to the model is 2000 bp. What happens if we only use the middle 200 bp for design?"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "a54c6ad3-f0f4-416c-a095-0725aec55b8c",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 0 -- Loss: 1.13, Improvement: N/A, Idx: N/A, Time (s): 0s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:00<00:00, 28.06it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 1 -- Loss: 0.657, Improvement: 0.4727, Motif Idx: 2, Pos Idx: 904, Time (s): 0.929\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:00<00:00, 29.36it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 2 -- Loss: 0.6221, Improvement: 0.03495, Motif Idx: 15, Pos Idx: 1099, Time (s): 0.8877\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:00<00:00, 28.05it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 3 -- Loss: 0.5962, Improvement: 0.02583, Motif Idx: 1, Pos Idx: 999, Time (s): 0.93\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:00<00:00, 29.01it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 4 -- Loss: 0.5871, Improvement: 0.00909, Motif Idx: 5, Pos Idx: 922, Time (s): 0.9012\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:00<00:00, 28.55it/s]"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 5 -- Loss: 0.582, Improvement: 0.005167, Motif Idx: 8, Pos Idx: 945, Time (s): 0.9153\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"y = predict(model, X)\n",
"y[:, idxs] = 3\n",
"\n",
"input_mask = torch.zeros(2000, dtype=bool)\n",
"input_mask[900:1100] = True\n",
"\n",
"X_hat2 = greedy_substitution(model, X, motifs, y, max_iter=5, input_mask=input_mask, verbose=True)"
]
},
{
"cell_type": "markdown",
"id": "eb99ee97-4aa6-4fc9-8927-3f9590b288b5",
"metadata": {},
"source": [
"As you might expect, we see a pretty big speed improvement -- around a 10x improvement in speed when only considering one tenth of the positions. That makes sense. But how do the results look?"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "34db6863-b9b7-414b-923b-d762b8bf27be",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"y0 = predict(model, X).numpy(force=True)[0]\n",
"y1 = predict(model, X_hat2).numpy(force=True)[0]\n",
"\n",
"plt.plot([-9, 5], [-9, 5], c='0.3')\n",
"plt.scatter(y0, y1, s=1)\n",
"plt.scatter(y0[idxs], y1[idxs], s=2, color='r', label='AP-1 Targets')\n",
"plt.legend()\n",
"plt.xlabel(\"Task Logits Before Design\")\n",
"plt.ylabel(\"Task Logits After Design\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"id": "962201b0-af02-4360-944e-940e48064754",
"metadata": {},
"source": [
"The results look quite similar to before. However, this is likely because the original sequence we used was completely randomly generated and the improvements are likely coming from simply adding AP-1 motifs nearby to each other, so limiting the possible space isn't a huge constraint. In settings with actual structured sequences as input or where running more iterations to get the maximum possible improvement is necessary, having more sequence can be helpful."
]
},
{
"cell_type": "markdown",
"id": "0c20d7d3-84ea-4832-9e4a-bc29af0cabaf",
"metadata": {},
"source": [
"#### Loss Functions\n",
"\n",
"By default, the loss used is `torch.nn.MSELoss` without reduction because it is a reasonable and general-purpose loss. However, you can use whatever loss you'd like by passing the loss function into `loss=...`. This can be any loss implemented in PyTorch that is reasonable for your model or any custom function with the signature `def loss(y, y_hat)` with `y.shape == (n, ...)` that returns a tensor with size `(n, ...)`. Basically, every example that gets passed into the loss function needs to have one or more losses associated with it. Everything except for the first dimension gets averaged such that you get one value per example. This is a critical detail because, by default, PyTorch losses will average over the batch dimension as well and only give you a single number regardless of the number of examples passed in. The design function needs to know the example with the minimal loss and so any function -- custom or not -- needs to return the loss for each example. \n",
"\n",
"To demonstrate, let's do the design but use a different loss function."
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "71bf8273-5963-4f2f-bad0-cc9710f23b36",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 0 -- Loss: 0.1229, Improvement: N/A, Idx: N/A, Time (s): 0s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.97it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 1 -- Loss: 0.1188, Improvement: 0.004089, Motif Idx: 5, Pos Idx: 910, Time (s): 8.753\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.97it/s]"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 2 -- Loss: 0.1185, Improvement: 0.0003167, Motif Idx: 14, Pos Idx: 692, Time (s): 8.774\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"y = predict(model, X)\n",
"y[:, idxs] = 3\n",
"\n",
"X_hat3 = greedy_substitution(model, X, motifs, y, max_iter=3, loss=torch.nn.HuberLoss(reduction='none'), verbose=True)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "1c838b7b-f789-44d6-8658-690d6a5acec7",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"y0 = predict(model, X).numpy(force=True)[0]\n",
"y1 = predict(model, X_hat2).numpy(force=True)[0]\n",
"\n",
"plt.plot([-9, 5], [-9, 5], c='0.3')\n",
"plt.scatter(y0, y1, s=1)\n",
"plt.scatter(y0[idxs], y1[idxs], s=2, color='r', label='AP-1 Targets')\n",
"plt.legend()\n",
"plt.xlabel(\"Task Logits Before Design\")\n",
"plt.ylabel(\"Task Logits After Design\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"id": "15c6d933-b6fd-4a9d-834f-ccc5f30401c4",
"metadata": {},
"source": [
"#### Elimination of Motifs\n",
"\n",
"An important part of this greedy procedure is that it does not only involve inserting new motifs into uninformative sequence, but that it will eliminate motifs that yield activity that the model does not want. As an illustration, let's consider the simple case where a motif drives unwanted AP-1 binding activity and the goal is to design a sequence that does not exhibit AP-1 binding.\n",
"\n",
"Here, we will start with a sequence that has the AP-1 motif in it already."
]
},
{
"cell_type": "code",
"execution_count": 15,
"id": "660a4744-40bd-4b17-ae7b-111e98325c2a",
"metadata": {},
"outputs": [],
"source": [
"from tangermeme.ersatz import substitute\n",
"\n",
"X = random_one_hot((1, 4, 2000), random_state=0).type(torch.float32)\n",
"X = substitute(X, \"GTGACTCATC\")"
]
},
{
"cell_type": "markdown",
"id": "b79852a6-5b03-492b-912e-e2ba18161c47",
"metadata": {},
"source": [
"Now, we will specify that we want baseline levels of AP-1 binding while preserving everything else."
]
},
{
"cell_type": "code",
"execution_count": 16,
"id": "b20f862c-2c62-4a09-afd4-e9ae7c3aa9bb",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 0 -- Loss: 0.1575, Improvement: N/A, Idx: N/A, Time (s): 0s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 2.99it/s]"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 1 -- Loss: 0.1182, Improvement: 0.03931, Motif Idx: 25, Pos Idx: 986, Time (s): 8.692\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"y = predict(model, X)\n",
"y[:, idxs] = -5\n",
"\n",
"X_hat3 = greedy_substitution(model, X, motifs, y, max_iter=1, verbose=True)"
]
},
{
"cell_type": "markdown",
"id": "0befbb89-5ed2-4336-8a09-77eed8dd87b2",
"metadata": {},
"source": [
"It looks like the process figured out that it should substitute in a motif that the model does not respond to (motif index 7) right in the middle of the sequence, where our AP-1 motif is, to eliminate it. Keep in mind that the process did not need to be told to eliminate motifs, or what motifs are uninformative. By being a comprehensive process, it can figure out both of these things in a data-driven manner. Further, the process can interleave the adding and eliminating of motifs -- it is not a \"one or the other\" situation. If some motifs need to be eliminated, and others need to be added, this procedure can figure that out (subject to the limitations of it being a greedy algorithm)."
]
},
{
"cell_type": "code",
"execution_count": 17,
"id": "379755bc-7f5e-471d-9898-7cf7ddc7c584",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"y0 = predict(model, X).numpy(force=True)[0]\n",
"y1 = predict(model, X_hat3).numpy(force=True)[0]\n",
"\n",
"plt.plot([-9, 5], [-9, 5], c='0.3')\n",
"plt.scatter(y0, y1, s=1)\n",
"plt.scatter(y0[idxs], y1[idxs], s=2, color='r', label='AP-1 Targets')\n",
"plt.legend()\n",
"plt.xlabel(\"Task Logits Before Design\")\n",
"plt.ylabel(\"Task Logits After Design\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"id": "e12f20ca-bb3f-46fe-876e-0b49501005ff",
"metadata": {},
"source": [
"And it looks like the predicted signal is lower after the elimination of the motif!\n",
"\n",
"We can do a more targetted approach by using the greedy saturated mutagenesis approach."
]
},
{
"cell_type": "code",
"execution_count": 18,
"id": "0413b121-b4f1-4c91-b33b-1e31facd17dd",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 0 -- Loss: 0.1575, Improvement: N/A, Idx: N/A, Time (s): 0s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|█████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:01<00:00, 2.95it/s]"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 1 -- Loss: 0.1187, Improvement: 0.03884, Motif Idx: 3, Pos Idx: 995, Time (s): 1.359\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"X_hat3 = greedy_substitution(model, X, ['A', 'C', 'G', 'T'], y, reverse_complement=False, max_iter=1, verbose=True)\n",
"\n",
"y0 = predict(model, X).numpy(force=True)[0]\n",
"y1 = predict(model, X_hat3).numpy(force=True)[0]\n",
"\n",
"plt.plot([-9, 5], [-9, 5], c='0.3')\n",
"plt.scatter(y0, y1, s=1)\n",
"plt.scatter(y0[idxs], y1[idxs], s=2, color='r', label='AP-1 Targets')\n",
"plt.legend()\n",
"plt.xlabel(\"Task Logits Before Design\")\n",
"plt.ylabel(\"Task Logits After Design\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"id": "156dac71-02d5-43f8-bc6c-3c437e6d837c",
"metadata": {},
"source": [
"With only a single mutation knocking out a key position in the motif, we are able to achieve similar results. "
]
},
{
"cell_type": "markdown",
"id": "994ea15b-c972-484d-b176-867e2a4dd214",
"metadata": {},
"source": [
"#### Half Precision\n",
"\n",
"A strength of this prediction-based design is that it is fairly robust to operating at lower precision because we are simply looking for the best motif and position, rather than needing a very precise value, and if two choices happen to be similar and swapped at lower precision the consequences for design are minimal."
]
},
{
"cell_type": "code",
"execution_count": 19,
"id": "1a582771-5e8b-49a0-828c-fc53e114b4f9",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Full Precision\n",
"Iteration 0 -- Loss: 28.47, Improvement: N/A, Idx: N/A, Time (s): 0s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:09<00:00, 2.89it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 1 -- Loss: 7.546, Improvement: 20.93, Motif Idx: 15, Pos Idx: 979, Time (s): 9.007\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 3.02it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 2 -- Loss: 4.787, Improvement: 2.759, Motif Idx: 15, Pos Idx: 1014, Time (s): 8.602\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:08<00:00, 3.04it/s]\n",
"/users/jmschr/anaconda3/lib/python3.12/site-packages/torch/nn/functional.py:3366: UserWarning: Applying the CPU mse kernel on half-type tensors. This may be slower than using float or double-type tensors. (Triggered internally at /opt/conda/conda-bld/pytorch_1716905971214/work/aten/src/ATen/native/cpu/BinaryOpsKernel.cpp:952.)\n",
" return torch._C._nn.mse_loss(expanded_input, expanded_target, _Reduction.get_enum(reduction))\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 3 -- Loss: 4.134, Improvement: 0.652, Motif Idx: 22, Pos Idx: 1040, Time (s): 8.564\n",
"\n",
"Half Precision\n",
"Iteration 0 -- Loss: 28.47, Improvement: N/A, Idx: N/A, Time (s): 0s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:05<00:00, 4.71it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 1 -- Loss: 7.543, Improvement: 20.92, Motif Idx: 15, Pos Idx: 979, Time (s): 5.53\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:05<00:00, 4.85it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 2 -- Loss: 4.789, Improvement: 2.754, Motif Idx: 15, Pos Idx: 1014, Time (s): 5.368\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:05<00:00, 4.84it/s]"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 3 -- Loss: 4.133, Improvement: 0.6562, Motif Idx: 22, Pos Idx: 1040, Time (s): 5.373\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n"
]
}
],
"source": [
"y = torch.zeros(1, 2002)\n",
"y[:, idxs] = 3\n",
"\n",
"print(\"Full Precision\")\n",
"X_hat = greedy_substitution(model.float(), X.float(), motifs, y.float(), output_mask=idxs, max_iter=3, verbose=True)\n",
"\n",
"print(\"\\nHalf Precision\")\n",
"X_hat = greedy_substitution(model.half(), X.half(), motifs, y.half(), output_mask=idxs, max_iter=3, verbose=True)"
]
},
{
"cell_type": "markdown",
"id": "38da5da9-ad79-4638-a733-165bd2f353f7",
"metadata": {},
"source": [
"Looks like we are getting the same motifs inserted at the same positions, but with a nice speed boost. Using half precision also potentially enables one to use a bigger batch size, which can be helpful for the modern GPUs that might be underutilized."
]
},
{
"cell_type": "markdown",
"id": "daf8da91-c14a-43e4-ba43-3c7b31054ddb",
"metadata": {},
"source": [
"### Designing Constructs\n",
"\n",
"An alternate view of design is that, rather than designing individual sequences, you want to design a construct with desired changes in model predictions *on average* when included in a sequence. Basically, rather than ending up with a single sequence where the design may have picked up on specific properties or other motifs already present, you want to design a small construct (of variable width) that can be included in other sequences to induce desired properties. This perspective is similar to the idea of de novo motif discovery, where one wants to find repeating patterns that *on average* are picked up on by the statistical or machine learning model.\n",
"\n",
"This method works by using repeated marginalizations for design. First, each individual motif is implanted into the middle of a set of background sequences and the one that pushes model predictions in the right direction the most is kept. Then, each motif is considered at each position from a provided `max_spacing` to the left of the left-most part of the motif to that same distance to the right. To be clear, when a `max_spacing` of 4 is provided, there will be a maximum of 4 positions between the right-most of the newly inserted motif and the left-most of the previous left flank, and likewise at the end. Because all positions in this span are considered, motifs can be implanted on top of each other to attenuate their effect. At the end of this process, you will end up with a construct that, on average, pushes model predictions by the amount provided as `y`. This is in contrast to the `greedy_substitution` method which will end up with a sequence where the total model predictions are `y`. \n",
"\n",
"On average, this function is roughly the same speed as the `greedy_substitution` function because, while it does not consider all positions in the sequence, it performs an in silico marginalization at each step over the number of provided background sequences. \n",
"\n",
"We can do this with a very similar function call. Remember here that `X` is a set of background sequences, not a single sequence to optimize, and that `y` is how much including this construct pushes model predictions, not the actual predictions of the model. Put another way, `y` is `f(X + construct) - f(X)`, not `f(X + construct)`."
]
},
{
"cell_type": "code",
"execution_count": 20,
"id": "17f1ecbb-98ee-4645-ba53-d7424c4e99b4",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 0 -- Loss: 1.0, Improvement: N/A, Idx: N/A, Time (s): 0s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:00<00:00, 70.07it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 1 -- Loss: 0.8707, Improvement: 0.1293, Motif Idx: 23, Pos Idx: 999, Time (s): 0.3736\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:09<00:00, 2.61it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 2 -- Loss: 0.4476, Improvement: 0.4231, Motif Idx: 25, Pos Idx: 983, Time (s): 9.956\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:12<00:00, 2.04it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 3 -- Loss: 0.3146, Improvement: 0.1329, Motif Idx: 22, Pos Idx: 997, Time (s): 12.82\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:13<00:00, 1.93it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 4 -- Loss: 0.2479, Improvement: 0.06674, Motif Idx: 25, Pos Idx: 1007, Time (s): 13.49\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:13<00:00, 1.89it/s]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Iteration 5 -- Loss: 0.1982, Improvement: 0.04969, Motif Idx: 8, Pos Idx: 1024, Time (s): 13.77\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████| 26/26 [00:17<00:00, 1.51it/s]\n"
]
},
{
"data": {
"text/plain": [
"torch.Size([4, 51])"
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from tangermeme.design import greedy_marginalize\n",
"\n",
"y = torch.zeros(1, 2002)\n",
"y[:, idxs] = 1\n",
"\n",
"X = random_one_hot((20, 4, 2000), random_state=0).float()\n",
"X_hat4 = greedy_marginalize(model, X, motifs, y, output_mask=idxs, max_iter=20, verbose=True)\n",
"X_hat4.shape"
]
},
{
"cell_type": "markdown",
"id": "902d4d85-6833-4cde-897f-4df2aa84465c",
"metadata": {},
"source": [
"A one-hot encoding is returned from this function, so we can take a look at what was designed using the `characters` function which is the conceptual opposite of a one-hot encoding."
]
},
{
"cell_type": "code",
"execution_count": 21,
"id": "75a74843-4501-4789-a01e-c2f90dd2b95c",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"'GGCGCGTGCCNNNNCTAAAAATAGGGCGCGTGCCNNNNNNNCGTCTAGACA'"
]
},
"execution_count": 21,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from tangermeme.utils import characters\n",
"\n",
"characters(X_hat4, allow_N=True)"
]
},
{
"cell_type": "markdown",
"id": "4863e3f8-26de-43da-8e02-5517816115fd",
"metadata": {},
"source": [
"Interestingly, adding in a single AP-1 motif would increase predictions by significantly more than the desired value of 1. Instead, the method constructs weaker versions of the binding sites by overlapping motifs."
]
},
{
"cell_type": "code",
"execution_count": 22,
"id": "5864988d-5a07-409a-a2a9-3762e3b7aec4",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"tensor(0.9614, dtype=torch.float16)"
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from tangermeme.marginalize import marginalize\n",
"\n",
"y_before, y_after = marginalize(model, X, X_hat4.unsqueeze(0))\n",
"(y_after - y_before)[:, idxs].mean()"
]
},
{
"cell_type": "markdown",
"id": "d3d6482a-5cc0-49eb-a2b6-677e400144f9",
"metadata": {},
"source": [
"#### Final Words\n",
"\n",
"Design can be a tricky process with many tradeoffs, particularly when it comes to genomics. One may need multiple iterations and trying out several strategies before converging on one that yields consistently good edits. This is, in part, because we have not yet fully converged on language for describing the pros and cons of various design methods. It is my hope that packages like `tangermeme`, and the design methods implemented here, can help move the field forward by standardizing some of the language we use to describe the design process."
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.9.12"
}
},
"nbformat": 4,
"nbformat_minor": 5
}