Official websites use .gov

A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS

A lock ( 🔒 ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Need Help

Towards a multimodal model of neuronal diversity in the human cerebral cortex

Award Number: F31MH146127
ORGANIZATION: NATIONAL INSTITUTE OF MENTAL HEALTH
OPDIV: NIH
AWARD CLASS: DISCRETIONARY
AWARD ACTIVITY TYPE: FELLOWSHIP/SCHOLARSHIP/STUDENT LOANS
PERIOD OF PERFORMANCE START DATE: 02/02/2027
PERIOD OF PERFORMANCE END DATE: 02/01/2028

Group Awards By:

View Award Description

Towards a multimodal model of neuronal diversity in the human cerebral cortex - PROJECT SUMMARY/ABSTRACT Classifying the approximately 16 billion neurons in the human cerebral cortex into discrete cell types with defined properties remains challenging. This work is crucial and would shed light on mechanisms of diseases with cell type specificity and identify cellular targets for brain-computer interfaces (BCIs) that can restore brain function in neurological disorders. Recent advances in high-throughput transcriptomics have revealed extensive molecular heterogeneity across human cortical neurons (hCNs), but gene expression alone cannot fully resolve CN types that differ in physiology or selective vulnerability. Thus, our lab pioneered Patch-seq to examine correlated variability across electrophysiologic, transcriptomic, and morphologic profiles paired at the single-neuron level. However, morphology remains the rate-limiting modality: reconstruction requires extensive manual post-processing, existing tools do not generalize well to human tissue, and downstream analyses rely on hand-engineered features with limited interpretability. This proposal aims to develop automated, interpretable machine learning (ML) tools for reconstructing and analyzing hCN morphology, and to integrate these data with existing transcriptomic datasets to discover multimodal hCN types. Aim 1 will build a computational toolkit to denoise, segment, and extract data-driven morphometrics from hCNs using convolutional neural networks (CNNs). I will test the hypothesis that learned representations capture biologically meaningful variability across hCNs, potentially identifying disease-relevant motifs missed by manual feature engineering. Aim 2 will leverage representation learning with large-scale human transcriptomic datasets to integrate modalities in human Patch-seq data, validating the robustness and discreteness of putative hCN types. Completing these aims will produce a scalable pipeline for human neuron reconstruction, a principled multimodal framework for defining neuronal types, and new insights into potential mechanisms of diseases mediated by specific hCN types. My primary sponsor, Cathryn Cadwell, MD PhD, is a neuropathologist-neuroscientist who developed Patch-seq and is an expert in the study of cortical circuitry. My co-sponsor, Vikaas Sohal, MD PhD, is a psychiatrist-neuroscientist with over a decade of expertise identifying the neuropsychiatric implications of dysfunction in specific hCN types. Their combined experience and commitment to my training will ensure that I learn to formulate and investigate biomedically relevant scientific questions, complete the proposed research, and gain the research and career development skills needed to achieve my long-term career goals. An F31 NRSA fellowship would allow me to deepen my expertise in integrating molecular, cellular, and computational neuroscience. This training will provide the foundation for me to pursue a career as an academic neurologist- neuroscientist, modeling the cellular architecture of the human brain to advance treatments for neurological disorders and inform cellular targets for new therapies and BCIs.


Issue Date FY	Funding FY	Legal Entity Name	Legal Entity Address	Legal Entity City	Legal Entity State	Legal Entity Zip Code	Legal Entity COUNTY	Legal Entity COUNTRY	Assistance Listing	Award Code	Budget Year	Action Date	Action Type	Action Amount

Issue Date FY: 2026 ( Subtotal = $43,479 )
2026	2026	REGENTS OF THE UNIVERSITY OF CALIFORNIA, SAN FRANCISCO, THE	1855 FOLSOM ST STE 425	SAN FRANCISCO	CA	94103	SAN FRANCISCO	USA	Mental Health Research Grants	000	1	8/5/2026	NEW	$43,479
														Subtotal = $43,479

Grand Total All Awards = $43,479

Top

All Categories

About

Search

Reports

Data Submission

Award Information

Towards a multimodal model of neuronal diversity in the human cerebral cortex

Award Number: F31MH146127

ORGANIZATION: NATIONAL INSTITUTE OF MENTAL HEALTH

OPDIV: NIH

AWARD CLASS: DISCRETIONARY

AWARD ACTIVITY TYPE: FELLOWSHIP/SCHOLARSHIP/STUDENT LOANS

PERIOD OF PERFORMANCE START DATE: 02/02/2027

PERIOD OF PERFORMANCE END DATE: 02/01/2028

Federal Websites

Department of Health & Human Services

HHS Operating Divisions

HHS Staff Divisions

Download A Document Viewer