# Advancing Homology Models to Identify Compounds for Understudied GPCRs

> **NIH NIH F32** · UNIVERSITY OF CALIFORNIA, SAN FRANCISCO · 2021 · $66,390

## Abstract

Project Summary/Abstract
Structure-based discovery is a proven method for identifying novel compounds that are potent and selective for
a given drug target. However, as this method relies on the knowledge of a drug target’s structure, we are limited
by the availability of known structures. G-protein coupled receptors (GPCRs) are the most heavily targeted family
of proteins, and yet, due to their nature as membrane proteins, our structural knowledge of these proteins is
limited. In order to leverage structure-based discovery for understudied GPCRs, we must use homology models
for virtual screens. There are examples of success when using homology models for prospective screens, but
there are many cases that have failed with causes unknown. In order to devise a set of rules that can more
effectively guide the use of homology models in virtual screens we must analyze each input to the modeling and
docking process systematically. Important unknowns include the accuracy of a homology model with respect to
a given crystal structure, the role of known actives, both in number and activity, that can be used to discriminate
effective models, and the ability to automate hit picking after a 400 million compound screen. It is expected that
slight deviations from a crystal structure may be helpful for virtual screens as a crystal structure is a
conformational snapshot, however it is unknown how much deviation is acceptable before prospective screens
fail. Further, selecting models based on their ability to successfully dock known agonists or antagonists could
enable prospective virtual screens to not only identify hits but identify hits with a given activity. Lastly, the ability
to automate much of the process would enable wide-scale application of the method to the large number of
understudied GPCRs and other proteins considered part of the dark genome. An important, immediate target for
application of these methods are in the identification of non-opioid receptor targeting therapies for pain
management. A family of GPCRs that may regulate pain pathways are the RF-amide peptide receptors.
Importantly, we have found that the pyroglutamylated RF-amide peptide receptor (QRFPR) is upregulated in
sensory neurons in a mouse model of chronic pain. Unfortunately, selective, small molecules have not been
made publicly available for QRFPR or other RF-amide receptors. The identification of tool compounds for these
receptors would allow researchers to effectively study the contributions of each receptor in the biology of pain.
Further, success in this application would provide more confidence in a systematic application of the method
towards the remaining understudied GPCRs.

## Key facts

- **NIH application ID:** 10084705
- **Project number:** 5F32GM136062-02
- **Recipient organization:** UNIVERSITY OF CALIFORNIA, SAN FRANCISCO
- **Principal Investigator:** Brian J Bender
- **Activity code:** F32 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2021
- **Award amount:** $66,390
- **Award type:** 5
- **Project period:** 2020-01-01 → 2021-12-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10084705

## Citation

> US National Institutes of Health, RePORTER application 10084705, Advancing Homology Models to Identify Compounds for Understudied GPCRs (5F32GM136062-02). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10084705. Licensed CC0.

---

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