Vikas Pejaver
Icahn School of Medicine at Mount Sinai, NY·Genetics
A single changed letter in a person's DNA can cause a rare disease or raise the risk of a common one. Working out which changes matter, and what to do about them, is still hard.
- Fields
Genetics
Computer Science
Medicine & Clinical Research
- Topics
variant interpretation
variant pathogenicity prediction
variant impact prediction
- Grants
- 2 known grants, through Jan 2029
What they work on
- Apply the team's previously developed variant impact predictors to large NIH Common Fund genomic data sets, such as Gabriella Miller Kids First and the Undiagnosed Disease Network, to identify known and predicted disease-associated variants.
- Select variants that directly or indirectly change druggable biochemical properties of proteins. Integrate them with drug data from programs such as LINCS and Illuminating the Druggable Genome to infer new drug-disease pairs for repurposing.
- Adapt and calibrate existing machine-learning pathogenicity predictors so that, for a specific gene, their scores become reliable estimates of whether a variant causes disease.
- Develop exomic disease risk scores built from pathogenicity predictor outputs, so that rare variants in protein-coding DNA count toward individual risk estimates, using data sets such as UK Biobank, dbGaP and Mount Sinai's BioMe.
- Integrate pathogenicity predictors into genome-wide polygenic risk scores, which sum the effects of many variants, and test whether this yields more accurate and clinically interpretable disease risk predictions.
More about the research
Variant pathogenicity predictors are machine-learning programs trained on known harmful and harmless DNA changes, and they can flag likely culprits. This work asks how these predictions can be pushed further in two directions: finding existing drugs that could treat rare genetic diseases, and turning predictions into accurate, clinically usable estimates of a person's disease risk.
Why it matters
Most rare genetic diseases have no targeted drug. Patients could benefit if existing drugs that block overactive proteins are matched to the diseases those proteins cause. In addition, calibrated variant predictors could let clinicians fold rare DNA changes into single-gene and whole-genome disease risk estimates in a way that fits medical guidelines. That could make genetic testing results more accurate and clinically useful.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Sai Zhang
Yale University, CT·Genetics
Develop a sequence-based deep learning model, trained on single-cell and individual-level data, that predicts gene expression from DNA sequence and estimates how genetic variants alter gene activity in each cell type.
Ariadne Letra
University of Pittsburgh, PA·Genetics
Determine the genetic architecture of familial cleft lip and palate (NSCLP) by combining linkage, case-control association and polygenic risk score analyses of whole genome sequences from 172 multigenerational families and additional datasets.
Eimear Kenny
Icahn School of Medicine at Mount Sinai, NY·Genetics
Develop robust clinical polygenic risk score (PRS) tests and combine them with standard clinical risk factors and family history to generate genomic risk assessments for up to 15 common diseases.
The school
Icahn School of Medicine at Mount Sinai: 45 labs on PI Pages, $1.0B a year on research, 62 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.