UNIVERSITY OF GENEVA · CENTER FOR INTEGRATIVE GENOMICS
Research Collaboration Proposal
Multi-ancestry rare disease genomics & regulatory variant interpretation
Dear Dr. Kapoor,
I hope this message finds you well. I am writing to explore a potential
research collaboration between our groups following your recent
presentation at the ISMB 2025 conference on deep learning
architectures for variant prioritization.
At the Center for Integrative Genomics, we have assembled a multi-
ancestry cohort of 4,200 whole genomes from families affected by rare
neurodevelopmental disorders, together with comprehensive RNA-seq
and methylation data from matched tissues. Your lab's work on
attention-based neural networks for regulatory region annotation —
particularly the Model-X architecture described in your
Genome
Research
paper last March — aligns closely with the analytical
challenges we face in interpreting non-coding variation.
I would like to propose a structured 18-month collaboration with two
concrete objectives. First, jointly develop an integrative scoring
framework that combines tissue-specific epigenomic signals with
population-level constraint metrics, implemented as an open-source
toolkit. Second, co-lead a collaborative R01 submission to the NIH
National Human Genome Research Institute (due June 15, 2026) to
scale the analysis to a clinically referred cohort of 12,000 additional
trios.
We would contribute the phenotyped cohort and wet-lab validation
pipeline; your group would lead the computational framework
development. We anticipate two cross-lab exchanges: one for your
postdoc to spend six weeks in our Geneva facility this autumn, and one
for our lead analyst to work from your Cambridge lab in early 2027.
Would you be available for a 30-minute video call in the coming two
weeks? I am free on Tuesday, May 27 between 14:00–17:00 CET, or
1 / 2
