# Charactering the impacts of regulatory epistasis with high-throughput precision genome editing.

> **NIH NIH F32** · STANFORD UNIVERSITY · 2020 · $64,926

## Abstract

PROJECT SUMMARY
Connecting genotype to phenotype remains a major challenge. Our inability to predict the fitness effects of
individual mutations, particularly when the effects of mutations are non-additive, have made it difficult to
connect genetic variation to phenotypes and disease traits. Changes in gene expression are thought to
frequently underlie phenotypic variation. However, how mutations within regulatory regions act to dictate
changes in gene expression is still not well understood. In particular, the role of additive versus non-additive
(epistatic) interactions between genetic variants within regulatory regions remains largely unexplored.
The proposed work will use an innovation in the CRISPR/Cas9 system (“CRISPEY”) to investigate the role of
epistasis in regulatory variation and evolution. CRISPEY greatly increases the efficiency of traditional precision
editing by generating a large number of potential donor DNAs in vivo using a bacterial retron element. In
preliminary analyses, this method shows ~100% efficiency of precision editing with no off-target edits. The first
CRISPEY scan assayed the fitness affects of 16,000 natural genetic variants differing between two strains of
Saccharomyces cerevisiae (RM and BY). Strikingly, it was found that the effects of proximal promoter variants
nearly always favored the same parental strain’s alleles. Reinforcement between variants in a cluster is not
expected under neutral evolution, and provides evidence of widespread lineage-specific selection acting on
promoter variants. Following this discovery, we will ask whether these proximal promoter variants affect fitness
additively or epistatically by creating combinatorial edits of variants in each cluster. First, we will characterize
general properties of regulatory epistasis. We will generate all possible combinations of 305 clusters of
promoter variants (n=5,392). The fitness of each combinatorial edit will be compared to that of the predicted
fitness based on individual variants to assess the extent, magnitude, and prevalence of different kinds of
epistasis. Second, we ask how natural selection shapes epistasis within regulatory regions by comparing
patterns observed in clusters of natural variants to a set of control variants. Next, we will ask how epistasis
constrains paths available for adaptive evolution of cis-regulation by assaying all possible paths between the
full BY and RM genotypes. Finally, we will ask whether epistasis for fitness results from epistasis for gene
expression levels. We will use qRT-PCR to quantify gene expression levels under different combinatorial edits
to assess the extent and magnitude of gene expression epistasis between natural variants. This will be the first
study to conduct a genome-wide survey of epistasis between natural variants within regulatory regions. This
work is critical to understanding how genetic variation translates to phenotypic variation, which is relevant for
understanding the genetic basis ...

## Key facts

- **NIH application ID:** 9934864
- **Project number:** 5F32GM131561-02
- **Recipient organization:** STANFORD UNIVERSITY
- **Principal Investigator:** Katya Mack
- **Activity code:** F32 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $64,926
- **Award type:** 5
- **Project period:** 2019-09-01 → 2022-08-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/9934864

## Citation

> US National Institutes of Health, RePORTER application 9934864, Charactering the impacts of regulatory epistasis with high-throughput precision genome editing. (5F32GM131561-02). Retrieved via AI Analytics 2026-05-22 from https://api.ai-analytics.org/grant/nih/9934864. Licensed CC0.

---

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