[ADD] scripts for the new analysis

jcapels · Oct 22, 2024 · 9a4e4d9 · 9a4e4d9
1 parent d0ffad5
commit 9a4e4d9
Show file tree

Hide file tree

Showing 28 changed files with 7,642 additions and 1,936 deletions.
diff --git a/examples/example.ipynb b/examples/example.ipynb
@@ -0,0 +1,86 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2024-09-17 13:55:42.809543: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
+      "2024-09-17 13:55:42.809613: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
+      "2024-09-17 13:55:42.811552: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
+      "2024-09-17 13:55:42.821527: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
+      "To enable the following instructions: AVX2 AVX512F FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
+      "2024-09-17 13:55:44.073516: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "WARNING:tensorflow:From /home/jcapela/miniforge3/envs/sm_precursor_predictor/lib/python3.10/site-packages/tensorflow/python/util/deprecation.py:588: calling function (from tensorflow.python.eager.polymorphic_function.polymorphic_function) with experimental_relax_shapes is deprecated and will be removed in a future version.\n",
+      "Instructions for updating:\n",
+      "experimental_relax_shapes is deprecated, use reduce_retracing instead\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Skipped loading modules with pytorch-lightning dependency, missing a dependency. No module named 'pytorch_lightning'\n",
+      "Skipped loading some Jax models, missing a dependency. No module named 'jax'\n",
+      "/home/jcapela/miniforge3/envs/sm_precursor_predictor/lib/python3.10/site-packages/deepmol/compound_featurization/__init__.py:20: UserWarning: Mol2Vec not available. Please install it to use it. (pip install git+https://github.com/samoturk/mol2vec#egg=mol2vec)\n",
+      "  warnings.warn(\"Mol2Vec not available. Please install it to use it. \"\n",
+      "LayeredFingerprint: 100%|██████████| 3/3 [00:00<00:00, 35.23it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "['', 'L-Proline;L-Arginine;L-Aspartate', 'L-Aspartate']\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sm_precursor_predictor import predict_precursors\n",
+    "precursors = predict_precursors(\n",
+    "            [\"CN1CCC=C(C1)C(=O)OC\", \"C1[C@H]2[C@@H](C[C@](N2)([C@H]([C@@H]1O)O)O)O\",\n",
+    "             \"C1CNCC(=C1)C(=O)O\"],\n",
+    "             model=\"Layered FP + Low Variance FS + Ridge Classifier\")\n",
+    "print(precursors)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "sm_precursor_predictor",
+   "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": "3.10.13"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/models_and_datasets/analysis/case_studies/predict_for_case_studies.ipynb b/models_and_datasets/analysis/case_studies/predict_for_case_studies.ipynb
diff --git a/models_and_datasets/analysis/challenging_datasets/assess_for_challenging_datasets.ipynb b/models_and_datasets/analysis/challenging_datasets/assess_for_challenging_datasets.ipynb
diff --git a/..._datasets/analysis/challenging_datasets/f1_scores_by_pipeline_and_precursor.png b/..._datasets/analysis/challenging_datasets/f1_scores_by_pipeline_and_precursor.png
diff --git a/models_and_datasets/analysis/challenging_datasets/legend.png b/models_and_datasets/analysis/challenging_datasets/legend.png
diff --git a/models_and_datasets/analysis/challenging_datasets/overlapping_dataset_results.png b/models_and_datasets/analysis/challenging_datasets/overlapping_dataset_results.png
diff --git a/...ts/analysis/challenging_datasets/precision_scores_by_pipeline_and_precursor.png b/...ts/analysis/challenging_datasets/precision_scores_by_pipeline_and_precursor.png
diff --git a/...asets/analysis/challenging_datasets/recall_scores_by_pipeline_and_precursor.png b/...asets/analysis/challenging_datasets/recall_scores_by_pipeline_and_precursor.png
diff --git a/models_and_datasets/analysis/dataset_analysis.ipynb b/models_and_datasets/analysis/dataset_analysis.ipynb
diff --git a/models_and_datasets/analysis/predictions_lotusdb/alkaloids_pie.png b/models_and_datasets/analysis/predictions_lotusdb/alkaloids_pie.png
diff --git a/models_and_datasets/analysis/predictions_lotusdb/glucosinolate.png b/models_and_datasets/analysis/predictions_lotusdb/glucosinolate.png
diff --git a/models_and_datasets/analysis/predictions_lotusdb/predict_for_lotus_sb.ipynb b/models_and_datasets/analysis/predictions_lotusdb/predict_for_lotus_sb.ipynb
diff --git a/models_and_datasets/analysis/results_for_alkaloids_dataset/pipeline_for_alkaloids_data.ipynb b/models_and_datasets/analysis/results_for_alkaloids_dataset/pipeline_for_alkaloids_data.ipynb
diff --git a/models_and_datasets/analysis/results_for_alkaloids_dataset/results_by_label_and_fold.csv b/models_and_datasets/analysis/results_for_alkaloids_dataset/results_by_label_and_fold.csv
diff --git a/models_and_datasets/analysis/results_for_alkaloids_dataset/significant_differences.csv b/models_and_datasets/analysis/results_for_alkaloids_dataset/significant_differences.csv
@@ -1,6 +1,6 @@
 Pipeline 1,Pipeline 2,Differences (F1 score)
-Layered FP + Low Variance FS + Ridge Classifier,Layered FP + Ridge Classifier,*
-Layered FP + Low Variance FS + Ridge Classifier,MGCNN (Eguchi et al. 2019),****
-Layered FP + Ridge Classifier,Morgan FP + Ridge Classifier,*
-Layered FP + Ridge Classifier,MGCNN (Eguchi et al. 2019),****
-Morgan FP + Ridge Classifier,MGCNN (Eguchi et al. 2019),***
+Layered FP + Ridge Classifier,NPClassifierFP + Ridge Classifier,*
+Layered FP + Ridge Classifier,MHFP + Ridge Classifier,****
+Layered FP + Ridge Classifier,Morgan FP + Ridge Classifier,****
+NPClassifierFP + Ridge Classifier,MHFP + Ridge Classifier,*
+NPClassifierFP + Ridge Classifier,Morgan FP + Ridge Classifier,**
diff --git a/...tasets/analysis/results_for_new_dataset/f1_scores_by_pipeline_and_precursor.png b/...tasets/analysis/results_for_new_dataset/f1_scores_by_pipeline_and_precursor.png
diff --git a/models_and_datasets/analysis/results_for_new_dataset/legend.png b/models_and_datasets/analysis/results_for_new_dataset/legend.png