Overview of DVApred

The above diagram depicts the schematic pipeline used in DVApred

The main contribution of this work is a manually curated database publicly shared, comprising of existing associations between viruses and their corresponding antivirals, obtained from DrugBank and other literature. The database gathers similarity information using the chemical structure of drugs by computaing the SIMCOMP score (using KEGG API) between them and the genomic sequence of viruses by finding d2* dissimilarity/distance (found using VirHostMatcher software). Along with this database, we make available a set of state-of-the-art computational drug re-positioning tools based on matrix completion.

Techniques used

The following computational algorithms based on matrix completion have been used to predict drug-virus associations :

  • Graph regularized matrix factorization [1]
  • Graph regularized matrix completion [2]
  • Graph regularized binary matrix completion [3]


  1. Ezzat, Ali, et al. "Drug-target interaction prediction with graph regularized matrix factorization." IEEE/ACM transactions on computational biology and bioinformatics 14.3 (2016): 646-656.
  2. Mongia, Aanchal, and Angshul Majumdar. "Drug-target interaction prediction using Multi Graph Regularized Nuclear Norm Minimization." Plos one 15.1
  3. (2020): e0226484.
  4. Mongia, Aanchal, Emilie Chouzenoux, and Angshul Majumdar. "Computational prediction of Drug-Disease association based on Graph-regularized one bit Matrix completion." bioRxiv (2020).