Software and Services

LeMeDISCO: An algorithm that predicts disease comorbidities from shared mode of action (MOA) proteins predicted by the AI-based MEDICASCY algorithm.

FINDSITEcomb: A tool for large scale virtual ligand screening. It offers the advantage that comparable results are obtained when predicted or experimental structures are used. The user can either provide a protein structure in PDB format or a protein sequence whose structure will first be predicted prior to its use in virtual ligand screening.
FINDSITEcomb2.0: An upgrade of the FINDSITE^comb algorithm that was compared to several commercially and freely available docking programs against the DUD set.
FRAGSITE: A fragment-based machine learning algorithm that scores protein-ligand binding for ligand virtual screening.
FRAGSITE2: A structure and fragment based approach for virtual ligand screening.
FRAGSITE^combM: A ligand virtual screening method that further improves to our previous best methods by combining the Morgan fingerprint (MF) with the originally used PubChem fingerprint and FP2 fingerprint.

ENTPRISE: An algorithm for predicting human disease-associated amino acid mutations from sequence entropy and predicted protein structures.
ENTPRISE-X: An algorithm for predicting human disease-associated frameshift & nonsense mutations.
DR. PRODIS: A comprehensive prediction of drug-protein interactions, side effects, toxicity and disease associations for the human proteome.
Know-GENE: A knowledge-based approach to predict gene–disease associations.
MEDICASCY: A multi-label based boosted random forest machine learning method that predicts small molecule’s side effects, indications, efficacy and mode of action proteins.
MOATAI-VIR: An AI algorithm that predicts severe adverse events and molecular features for COVID-19’s complications.
PHEVIR: An algorithm which predicts the corresponding human disease comorbidities of 312 viruses and 57 bacteria.
PICMOA: An algorithm that classifies 3,526 diseases across 20 clinically classified classifications into 26 molecular classes.

DESTINI: A deep-learning based contact-driven protein structure prediction tool.