# Signatures of interaction-driven selection on low-complexity sequences > **NIH NIH R01** · UNIVERSITY OF CHICAGO · 2021 · $277,548 ## Abstract Abstract Eukaryotic cells respond to stress (heat, oxidation, starvation) by reorganizing proteins and RNA into massive, reversible assemblies—termed stress granules—thought to play important roles in stress survival. Stress granules contain dozens of RNA-binding proteins, most of which contain disordered, quasi-repetitive low- complexity regions (LCRs). Recent, revolutionary work has demonstrated that these low-complexity regions mediate or modulate the formation of large assemblies by specific biophysical processes: phase separation and hydrogel formation. The resulting liquid/gel assemblies lack the fixed stoichiometry of quaternary structures and have been termed quinary assemblies. Disease-associated mutations in LCRs perturb quinary assemblies, producing pathological fibrillarization associated with these mutations in human disease. Yet the constraints on LCRs, and how LCR sequences encode quinary behaviors, remain unclear. Evolutionary analyses grounded in the biophysics of quinary assembly are urgently needed. How do low-complexity sequences encode quinary assembly behavior? Which features of these highly variable sequences are under selection, and which reflect mutational processes long known to give rise to repetitive low-complexity sequences? How can we uncover coevolution between features in interacting LCRs, in the absence of reliable site-specific information? And how can we experimentally validate putative evolutionary constraints? We propose an integrative approach to answering these questions. This approach builds on our deep expertise in analyzing evolutionary constraints on protein sequences, and in evolutionary and empirical studies of protein aggregation, coupled with our recent work identifying and isolating stress-triggered phase-separation behavior in RNA binding proteins. We have recently discovered that poly(A)-binding protein (Pab1 in budding yeast), an abundant, conserved RNA-binding protein with a highly variable LCR, phase-separates in response to heat and pH stress. We discovered novel patterns of evolutionary constraint in this LCR, and showed that making mutations which systematically perturbing the conserved composition of this LCR result in coordinately perturbed phase-separation, gel formation by Pab1, and altered stress survival by the organism. In Aim 1, we propose methods for quantifying these and related evolutionary constraints which generalizes broadly to sequences showing selection on amino-acid composition with weak or negligible selection on the ordering of amino acids. Based on substantial preliminary work, we propose specific methods to quantify selection linked to the biophysical features needed to promote phase separation while preventing pathological aggregation. In Aim 2, we describe an experimental system for assessing the in vitro and in vivo effects of perturbing evolutionarily constrained LCR features. In Aim 3, we propose new methods for quantifying covariation within and between LCRs. Toget... ## Key facts - **NIH application ID:** 10115071 - **Project number:** 5R01GM127406-04 - **Recipient organization:** UNIVERSITY OF CHICAGO - **Principal Investigator:** David Allan Drummond - **Activity code:** R01 (R01, R21, SBIR, etc.) - **Funding institute:** NIH - **Fiscal year:** 2021 - **Award amount:** $277,548 - **Award type:** 5 - **Project period:** 2018-05-01 → 2022-02-28 ## Primary source NIH RePORTER: https://reporter.nih.gov/project-details/10115071 ## Citation > US National Institutes of Health, RePORTER application 10115071, Signatures of interaction-driven selection on low-complexity sequences (5R01GM127406-04). Retrieved via AI Analytics 2026-05-22 from https://api.ai-analytics.org/grant/nih/10115071. Licensed CC0. --- *[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*