Deep learning methods for metabolite identification from high resolution LC-MS data.

Metabolomics, or the study of all cellular metabolites, promises to be a new cornerstone in disease diagnosis and precision medicine.  Several recent reports demonstrate that metabolite profiles are excellent indicators of the health status of individuals.  In addition, metabolomics monitoring is likely to be included in the clinical trials on a routine basis.  However, the field of metabolomics is not nearly as developed as proteomics or transcriptomics.  Identification of a large number of metabolites is a challenge.  Our goal will be to develop deep learning algorithms for LC-MS/MS data analysis; especially for metabolite identification.  Tools such as MetDNA and GNPS are available in the public domain and the student will first explore them and identify the areas of improvement.  Familiarity with the concepts of AI/ML and basic proficiency in programming is expected.

 

Co-guide:  Prof. Pushpak Bhattacharya, CSE Department.

 

UG Project Type
BTP
SLP
Name of Faculty