Biomol_classifier
the optimized algorithm is inside the forward__feature_selection folder, same for results and other data
cns_smiles.txt | original txt with cns smiles and names
non_cns_smiles.txt | original txt with non cns smiles and names
cns_smiles.txt and non_cns_smiles.txt were taken from: hose, A. K., Herbertz, T., Hudkins, R. L., Dorsey, B. D., & Mallamo, J. P. Knowledge-based, central nervous system (CNS) lead selection and lead optimization for CNS drug discovery. ACS Chemical Neuroscience, 3(1), pp 50–68, 2012.
cns_molecues.csv | only cns molecules with all features
non_cns_molecues.csv | only non ns molecules with all features
molecules.csv | all features
e_molecules.csv | experimental features
f_classif_em_molecules.csv | kbestfeatures after f_classif and data cleaning. Experimental features + must have features
f_classif_emp_molecules.csv | kbestfeatures after f_classif and data cleaning. Experimental features + must have features + possible features
f_classif_best.csv | best k features after f_classif cleaning. k=1 to k=all-1 as values
mic_best.cvs | best k features after mutual info classif cleaning. k=1 to k=all-1 as values