Merge pull request #4 from igoradaptyv/biolm/reproducibility-plots

Add BioLM EGFR analysis notebooks for ESM log-probabilities, DNA features, and molecule types

Author: igoradaptyv

biolm/.gitignore

@@ -0,0 +1,9 @@
+# Generated outputs
+plots/
+
+# Jupyter notebook checkpoints
+.ipynb_checkpoints/
+
+# Python cache
+__pycache__/
+*.pyc

biolm/README.md

@@ -0,0 +1,32 @@
+# BioLM EGFR Competition Analysis
+
+Analysis notebooks for protein language model log-probabilities, DNA feature importance, and molecule type comparisons.
+
+## Setup
+
+```bash
+cd biolm/
+python3 -m venv .venv
+source .venv/bin/activate
+pip install -r requirements.txt
+```
+
+## Generate Figures
+
+```bash
+cd biolm/
+
+# ESM log-probability analysis → plots/figure13_log_probabilities.png
+jupyter nbconvert --to notebook --execute protein_llm_log_probability_plot.ipynb
+
+# DNA feature importance → plots/figure18_feature_importance.png
+jupyter nbconvert --to notebook --execute dna_feature_importance_plot.ipynb
+
+# Molecule type ANOVA → plots/anova_molecule_types.csv
+jupyter nbconvert --to notebook --execute molecule_type_anova_table.ipynb
+
+# Molecule type distribution → plots/molecule_type_distribution.png
+jupyter nbconvert --to notebook --execute molecule_type_plot.ipynb
+```
+
+Data files in `data/` are tracked with Git LFS. If not installed: `brew install git-lfs && git lfs install` (macOS) or `sudo apt install git-lfs && git lfs install` (Ubuntu).

…m/egfr_adaptyv_affinity_biolm_silica.csv …a/egfr_adaptyv_affinity_biolm_silica.csvbiolm/egfr_adaptyv_affinity_biolm_silica.csv renamed to biolm/data/egfr_adaptyv_affinity_biolm_silica.csv biolm/egfr_adaptyv_affinity_biolm_silica.csv renamed to biolm/data/egfr_adaptyv_affinity_biolm_silica.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb14500d371c00adb07c9ca1ba97b4d2458daeb57cf44958d3ed08a885b674ae
+size 2271076