# Temporally resolved and interpretable machine learning model of GPCR conformational transition

## TL;DR

A new interpretable machine learning model, DRUMBEAT, identifies amino acid residues involved in protein conformational transitions to aid in designing subtype-specific drugs.

## Contribution

The novel DRUMBEAT algorithm enables interpretable and dynamic analysis of protein conformational transitions for drug design.

## Key findings

- DRUMBEAT identifies residue communities regulating conformational state ensembles in β2-adrenergic receptor.
- Distinct residue communities around specific contacts are found to be unique to D3R conformational transitions compared to D2R.
- These findings can guide the development of subtype-specific drugs for neuropsychiatric disorders.

## Abstract

Identifying target-specific drugs remains a challenge in pharmacology, especially for highly homologous proteins such as dopamine receptors D2R and D3R. Differences in target-specific cryptic druggable sites for such receptors arise from the distinct conformational ensembles underlying their dynamic behavior. While Molecular Dynamics (MD) simulations has emerged as a powerful tool for dissecting protein dynamics, the sheer volume of MD data requires scalable and unbiased data analysis strategies to pinpoint residue communities regulating conformational state ensembles. We present the Dynamically Resolved Universal Model for BayEsiAn network Tracking (DRUMBEAT) interpretable machine learning algorithm and validate it by identifying residue communities that enable the deactivation of the β2-adrenergic receptor. Further, upon analyzing dopamine receptor dynamics we identify distinct and non-conserved residue communities around the contacts F1704.62_F172ECL2 and S1464.38_G14134.56 that are specific to D3R conformational transitions compared to D2R. This information can be tapped to design subtype-specific drugs for neuropsychiatric and substance use disorders.

Differences among homologous receptor proteins complicate target-specific drug design. Here, authors develop an interpretable dynamic machine learning model DRUMBEAT to identify amino acid residues enabling distinct conformational transitions in proteins.