PREDICT: a method for inferring novel drug indications with application to personalized medicine

By Assaf Gottlieb, Gideon Y Stein, Eytan Ruppin, Roded Sharan. 2011

815 citations recorded by the source corpus.

Inferring potential drug indications, for either novel or approved drugs, is a key step in drug development. Previous computational methods in this domain have focused on either drug repositioning or matching drug and disease gene expression profiles. Here, we present a novel method for the large‐scale prediction of drug indications (PREDICT) that can handle both approved drugs and novel molecules. Our method is based on the observation that similar drugs are indicated for similar diseases, and utilizes multiple drug–drug and disease–disease similarity measures for the prediction task. On cross‐validation, it obtains high specificity and sensitivity (AUC=0.9) in predicting drug indications, surpassing existing methods. We validate our predictions by their overlap with drug indications that are currently under clinical trials, and by their agreement with tissue‐specific expression information on the drug targets. We further show that disease‐specific genetic signatures can be used to accurately predict drug indications for new diseases (AUC=0.92). This lays the computational foundation for future personalized drug treatments, where gene expression signatures from individual patients would replace the disease‐specific signatures.

Subfields: Computational Theory and Mathematics, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Molecular Biology

Citations in the corpus

The SIDER database of drugs and side effects (2015, methodology): “Thanks to the availability of downloadable files, SIDER has since been used in many studies, for example to identify metabolic dysregulation as a cause for ADRs (8), to investigate the effect of essential proteins on ADRs (9) and to predict drug indications (10).”
Network-based in silico drug efficacy screening (2016, neutral): “The increasing interest in reusing existing drugs for novel therapies has recently given rise to various approaches that aim to identify candidate drugs with similar characteristics to known drugs used in a disease 7,8,19–20 .”
Systematic integration of biomedical knowledge prioritizes drugs for repurposing (2017, methodology): “A 2011 study, named PREDICT, compiled 1933 treatments between 593 drugs and 313 diseases (Gottlieb et al., 2011).”
The Emerging Paradigm of Network Medicine in the Study of Human Disease (2012, background): “78 Consequently, these types of studies may form the computational foundation for future development of personalized therapies, in which variations in gene expression signatures and specific phenotypes from individual patients could be considered in order to dictate the course of care.”
Graph embedding on biomedical networks: methods, applications and evaluations (2019, positive): “drug-disease association graphs (Gottlieb et al., 2011), detecting long noncoding RNA (lncRNA) functions based on lncRNA-protein interaction networks (Zhang et al.”
Molecular Systems Biology Peer Review Process File Indi: a Computational Framework for Inferring Drug Interactions and Their Associated Recommendations Transaction Report (neutral): “For each query drug pair, we constructed features expressing its similarity to the closest gold-standard DDI, by adapting the scoring scheme of Gottlieb et al (2011) (Materials and methods).”
The Human Phenotype Ontology project: linking molecular biology and disease through phenotype data (2013, background): “…analysis of the distribution of proteins in the postsynaptic density of the human neocortex (31), to derive a disease–disease similarity measure for the prediction of novel drug indications (32) and to analyze overrepresentation of phenotypes associated with individual protein domains (33).”
Towards Personalized Medicine: Leveraging Patient Similarity and Drug Similarity Analytics (2014, background): “Drug similarity analytics has been widely used in drug repositioning 7-9 , drug side-effects prediction 10 , drug-target interactions prediction 11 , and drug-drug interactions prediction 12, 13 applications.”
Drug repositioning: a machine-learning approach through data integration (2013, background): “Finally, more comprehensive methods that take into account chemical, molecular and biological aspects of the drug-disease interactions have also been recently proposed [8].”
Artificial intelligence in cancer target identification and drug discovery (2022, methodology): “The ﬁ rst network-based biology analysis application of drug repurposing quanti ﬁ es the similarities or relationships for known drug-disease associations, and then uses regression models or statistical models to predict novel drug-disease associations 234,235 .”