# Cell Systems to Pre-Clinical Models

> **NIH NIH U54** · UNIVERSITY OF MICHIGAN AT ANN ARBOR · 2020 · $581,933

## Abstract

PROJECT SUMMARY
Knowledge of epilepsy genes has exploded with the advent of next generation DNA sequencing (NGS), yet
our understanding of how any specific gene mutation leads to epilepsy increasingly lags behind gene
discovery. With the expanding list of potential epilepsy genes comes the problem of genetic variants of
uncertain significance (VUS), confounding definitive diagnosis, genetic counselling and treatment strategies.
The ultimate goal of the EpiMVP team is to develop a highly integrated and effective multiplatform pipeline
to determine the functional impact of VUS for non-ion channel epilepsy genes. To this end, the Gene and
Variant Curation Core (GVCC) provides in silico prediction of VUSs to prioritize and refine specific genes
and VUS for study as Project 1 (Genes to Proteins) begins in vitro biochemical and cell biological
assessment and Project 2 (Proteins to Cell Systems) further explores the prioritized VUS in two
dimensional (2-D) and 3-D cell systems. However, due to the complex nature of epilepsy and the brain,
uncertainty and even conflicting interpretation will at times arise from in vitro biochemical and cell-based
functional assays. Thus, it is ultimately critical to apply refined VUS list for in vivo functional assessment in
model organisms. The long-term goal of project 3 (Cell Systems to Preclinical Models) is to incorporate
in vivo functional assays into EpiMVP pipeline in which the functional and genetic data of a specific VUS are
merged to arrive at a decisive probability of pathogenicity. Project 3 will take advantage of Clustered
Regularly Interspaced Short Palindromic Repeats (CRISPR) gene editing methods to generate well-defined
loss-of-function (LOF) epilepsy models in rodent and zebrafish as a null background on which to evaluate
human VUS. Data generated in Project 3 will be shared with Projects 1 and 2 and the GVCC in an iterative
fashion with an overall deliverable to develop EpiPred (Epilepsy Variant Prediction), a machine learning
model that will allow accurate classification of missense epilepsy gene variants as likely pathogenic or
benign. With the prioritized and refined VUS list, we will test 3-4 VUS each for 1-2 genes per year and
complete testing of at least 5 genes in both fish and rodent model systems. Our first-year goal is to examine
functional rescue of 3-4 VUS each for STXBP1 and DEPDC5. Deliverables include: 1) Cross-validated in
vivo models to interrogate epilepsy genes; 2) Determination of VUS pathogenicity for at least 5 non-ion
channel epilepsy genes using 2 in vivo models; 3) Assessment of in silico and in vitro models for VUS
pathogenicity prediction, and whether in vivo models are required for certain genes and VUS; and 4)
Optimized in vivo models for each epilepsy gene that will be powerful platforms for future drug discovery and
for exploring underlying mechanisms in these genetic epilepsies.

## Key facts

- **NIH application ID:** 10003683
- **Project number:** 1U54NS117170-01
- **Recipient organization:** UNIVERSITY OF MICHIGAN AT ANN ARBOR
- **Principal Investigator:** Yu Wang
- **Activity code:** U54 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $581,933
- **Award type:** 1
- **Project period:** 2020-09-15 → 2025-07-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10003683

## Citation

> US National Institutes of Health, RePORTER application 10003683, Cell Systems to Pre-Clinical Models (1U54NS117170-01). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10003683. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*