Back

Pathdict: Machine learning software predict Drug-Pathway interaction in human Based on drug simplified molecular-input line-entry system (SMILES)

Habib, P.; Alsamman, A. M.; Hassanein, S.; Hamwieh, A.

2020-01-09 bioinformatics
10.1101/2020.01.08.899005 bioRxiv
Show abstract

Predicting the target of unknown or/and drugs under investigation from data of already identified drugs is very important not only for the understanding of various drug and molecular interaction processes but also for the development of novel drugs. Here we introduce TarDict, a RandomForestClassifier based-software predict the target pathway or protein based on SMILES of chemical. TarDict receives SMILES and returns a list of the possible similar drug, then export list to the user the target that drug contribute in. Training data set of 20442 entry and testing reveal %95 accuracy.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.