Debarshi Mustafi
University of Washington, WA·Genetics·Early-career
Many inherited eye diseases and cancers remain hard to diagnose or predict. One reason is that standard DNA tests read only short stretches of DNA.
- Fields
Genetics
Medicine & Clinical Research
Computer Science
- Topics
retinoblastoma
tumor suppressor
hereditary cancer
- Grants
- 2 known grants, through Jul 2031
What they work on
- Use targeted long-read sequencing of RB1 DNA and methylation from a patient's sample alone to assign the parent of origin of the inherited variant, and link it to retinoblastoma severity, treatment response and second-cancer risk.
- Test whether an alternative RB1 transcript from the father's gene copy interferes with normal RB1 activity, using direct RNA and cDNA long-read sequencing to detect gene isoforms and parent-specific expression.
- Develop deep-learning neural networks that detect parent-specific DNA methylation across the genome, to find new methylated regions linked to other hereditary cancers.
- Find missing non-coding USH2A variants in Usher syndrome patients with only one known mutation, by combining Illumina short-read and Nanopore long-read sequencing to assemble and phase the gene region.
- Test whether candidate non-coding USH2A variants cause disease using luciferase reporter cells and patient-derived retinal organoids edited with CRISPR-Cas9.
More about the research
These short reads can miss changes outside protein-coding regions. They also cannot show which parent a variant came from, or whether chemical tags called methylation are present. Methylation is a chemical mark on DNA that helps switch genes on or off. The lab uses long-read sequencing, which reads much longer DNA or RNA molecules in one piece, to address these gaps. Its two projects study retinoblastoma, a childhood eye cancer, and Usher syndrome, an inherited cause of combined deafness and blindness.
Why it matters
Children with retinoblastoma and their families could benefit from a single test that predicts disease severity, treatment response and later cancer risk. The same test could point toward antisense-based therapies. Patients with suspected Usher syndrome who now lack a full genetic diagnosis could receive one. Identifying their non-coding variants could also make them candidates for gene-targeting or gene-editing therapies. More broadly, the methods could improve risk assessment for other hereditary cancers and inherited retinal diseases.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Yang Sun
Stanford University, CA·Genetics
Determine how mitochondrial proteins help position ARMC9, a Joubert syndrome gene, in photoreceptor cilia, to explain the newly found link between mitochondrial function and cilia building.
Gemma Carvill
Northwestern University, IL·Genetics
Identify the RNA-binding proteins that cause extra splicing-in of an SCN1A poison exon in people with epilepsy, using splice reporters and proteomics, and test whether antisense oligonucleotides correct it in patient iPSC-derived neurons.
John Edwards
Washington University in St Louis, MO·Neuroscience
Develop a nanopore long-read sequencing method that simultaneously and accurately profiles 5-methylcytosine and 5-hydroxymethylcytosine in both CG and non-CG contexts across gene promoters and gene bodies.
The school
University of Washington: 81 labs on PI Pages, $1.7B a year on research, 694 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.