Computational Biology

The Computational Biology group was established recently at MSMF with a mandate to utilize computational and statistical techniques including Artificial Intelligence (AI) and Machine Learning (ML) algorithms for processing clinical, molecular, and medical image data to generate hypotheses for validation and translation to the clinic as diagnostics, prognostics or therapeutics. Currently, the group is working on projects in the area of head and neck and oncology and neuro-oncology towards identifying novel biomarkers through integrating multi-omic data. In another realm of research, members of the computational biology group analyze images of routinely acquired histopathology slides to uncover hidden patterns that link to a clinically meaningful subtype in brain tumors or to therapy sensitivity in lung carcinoma. All these projects could potentially lead to a diagnostic or prognostic test in the clinic that can aid the clinicians in augmenting the disease management regime.

Dr. Sujan K Dhar

Research Projects

  • Proteogenomic Integration for discovery of differntial biomarkers in Glioblastoma: Glioblastoma is an aggressive form of brain cancer with median survival less than 18 months. In this project, we are analyzing transcriptomics and proteomics data generated from a set of glioblastoma and lower-grade glioma patients using a combination of de novo assembly and alignment-based pipelines to arrive at novel peptides differentially expressed in glioblastoma and other related clinical phenotypes. Dr Ravi Sirdeshmukh, Dr Abhishek Kumar, Institute of Bioinformatics.
  • AI-enabled computational model for IDH1 mutation detection from H&E-stained glioma histopathology images : Mutations in the IDH gene are known to indicate better prognosis for patients suffering from lower-grade glioma. Current pathology practice of detecting the mutation using immunohistochemistry can only detect the canonical mutations. In this project, we aim to develop an artificial neural network-based computational model that can detect both canonical and non-canonical mutations from histopathology images that are generated as part of the standard of care. Dr Akhila L, Pathology, NH
  • Determination of COVID-19 severity indicators from the systematic analysis of public data: Through successive waves of COVID-19, it is now well-established that around 5-10% of infected patients progress to severe form of the disease and other factors like age and pre-existing illnesses increases the chance of severity. In this study, we have analyzed a plethora of publicly available data to establish a set of cytokine markers like IL-6 and IL-10 that be used as a prognostic indicators. We also showed that inflammation and hypercoagulopathy are the pre-dominant risk factors of severity for SARS-CoV-2 infected patients suffering from diabetes.


  • Tumor microenvironment platform comprising oral carcinoma for predicting drug sensitivity, drug resistance, and disease progression. File Number: 202141048345/TEMP/E-1/54650/2021-CHE. Filed Date: 23RD OCTOBER 2021
  • Compositions and methods for treating coronavirus infection with different levels of disease severity. File Number: 202041036866. Filed Date: 27TH AUGUST 2020
  • Antibodies against Lipocalin-2 and uses thereof. File Number: 202021000274. Filed Date: 03RD JANUARY 2019