Re: [EXTERNAL] Re: Invitation to Science Forum Microsoft Research AI for Science

Title:

AI Approaches for Understanding Brain Disorders:
Schizophrenia, Alzheimer's disease & ADHD

Abstract:

My talk will focus on the material in:

Emani et al. (2024). Science.
Single-cell genomics and regulatory networks for 388 human brains

Specifically, I'll discuss:

Single-cell genomics is a powerful tool for studying heterogeneous
tissues such as the brain. Yet little is understood about how genetic
variants influence cell-level gene expression. Addressing this, we
uniformly processed single-nuclei, multiomics datasets into a resource
comprising >2.8 million nuclei from the prefrontal cortex across 388
individuals with various brain-related disorders and controls. For 28
harmonized neuronal and non-neuronal cell types, we assessed
population-level variation in expression and chromatin across gene
families and drug targets. Integration of expression and genotype data
revealed >1.4 million single-cell expression quantitative trait loci
(eQTLs), many of which were not seen in bulk gene-expression datasets.
The chromatin datasets allowed for the identification of >550,000
single-cell cis-regulatory elements enriched at loci linked to
brain-related traits. Combining expression, chromatin, and eQTL
datasets, we built cell type–specific gene regulatory and cell-to-cell
communication networks, which manifest cellular changes in aging and
neuropsychiatric disorders, including altered Wnt signaling in
schizophrenia and bipolar disorder. We further constructed an
integrative deep-learning model that accurately imputes single-cell
expression and simulates perturbations. The model prioritized ~250
disease-risk genes and drug targets with associated cell types,
suggesting potential precision-medicine approaches for
neuropsychiatric disorders.

If there's time, I'll also touch on the material in :

* A Variational Graph Partitioning Approach to Modeling Protein
Liquid-liquid Phase Separation
Gaoyuan Wang, Jonathan H Warrell, Suchen Zheng, Mark Gerstein (2023).
Cell Reports Physical Science.

* Leveraging a large language model to predict protein phase
transition: a physical, multiscale and interpretable approach
M Frank, P Ni, M Jensen, MB Gerstein (2024). Proc Natl Acad Sci U S A
121: e2320510121.

* Digital phenotyping from wearables using AI characterizes
psychiatric disorders and identifies genetic associations
JJ Liu, B Borsari, Y Li, SX Liu, Y Gao, X Xin, S Lou, M Jensen, D
Garrido-Martin, TL Verplaetse, G Ash, J Zhang, MJ Girgenti, W Roberts,
M Gerstein (2024). Cell.

i0msft