Vinicius Contessoto
Rice University, TX·Biophysics
Each chromosome is a very long DNA molecule folded inside the cell nucleus, and the way it folds helps decide which genes switch on. Some experiments average this folding over millions of cells.
- Fields
Biophysics
Physics
Genetics
- Topics
3D genome
chromosome structure
chromatin folding
- Grants
- 1 known grant, through Jun 2029
What they work on
- Develop ChromVAE, a variational autoencoder (a generative neural network) trained on high-resolution DNA-tracing data, to learn a compact map of chromatin folding and generate new chromosome structures consistent with experiments.
- Create the Associative Memory for Imaging (AMI) force field for maximum entropy polymer simulations, tuned to reproduce the distributions of distances between DNA regions measured by single-cell microscopy rather than average Hi-C contacts.
- Predict how genomic structural variants (deletions, duplications and inversions) change 3D genome folding, using force fields learned from normal (wild-type) data without additional parameter training.
More about the research
Microscopes that trace DNA in single cells show that each copy folds differently. This project builds physics-based computer models that capture that cell-to-cell variety. It then uses them to predict how DNA rearrangements change the genome's 3D shape and, potentially, gene regulation.
Why it matters
Genomic rearrangements such as deletions, duplications and inversions can change how the genome folds and may alter gene regulation. Models that predict these 3D changes directly from physics could help researchers interpret genomic datasets and judge the likely effects of genetic alterations. Open-source tools such as ChromVAE and OpenMiChroM would let other labs apply these models to new imaging data.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Michael Shirts
University of Colorado Boulder, CO·Chemistry
Build systematically improvable force fields with the Open Force Field Initiative, optimized and validated against new thermophysical data, with statistical tests of when added complexity improves accuracy.
Alexander D Mackerell
University of Maryland, Baltimore, MD·Biophysics
Refine the CHARMM36 force field, a model of atomic interactions used in simulations, by directly optimizing Lennard-Jones terms between atom pairs instead of relying on approximate combining rules.
Alistair Boettiger
Stanford University, CA·Biophysics
Measure how genomic distance and context shape 3D-genome motion, meaning the frequency, speed and timescales of contacts, at a simple and a complex locus using TRACK-IT live imaging with transposon-tiled fluorescent DNA labels.
The school
Rice University: 24 labs on PI Pages, $278M a year on research, 267 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.