Biology subjects

Good, L. L.

Publications and source records attributed to Good, L. L..

4 recordsLinked to original sources

bioRxiv2026

Mechanistic Interpretability of Protein Language Models Reveals Encoded Structural and Functional Properties of Intrinsically Disordered Proteins

Protein language models (PLMs) such as ESM-2 encode protein sequences as embeddings for downstream tasks. PLMs are trained on a masked learning objective that leverages evolutionary constraints. While interpretability studies of ESM-2 have focused on folded proteins, their behavior on intrinsically disordered proteins (IDPs), which constitute a substantial fraction of the human proteome and are implicated in numerous diseases, remains understudied. Because IDPs experience different types of evolutionary constraints on their amino acid sequences, we hypothesized that PLMs would behave differently on disordered versus folded regions. Here we show that ESM-2 exhibits reduced attention on disordered regions, yet still encodes meaningful biological signals. The model assigns heightened attention to disease-relevant residues even at high levels of disorder. Moreover, we show that both the radius of gyration and individual dynamic contact maps, key characteristics of IDPs, can be obtained from the model logits and embeddings. These findings suggest PLMs capture valuable information relevant to IDP biology despite their bias toward structured residues.

bioinformatics↗

bioRxiv2024

Mechanical Profiling of Biopolymer Condensates through Acoustic Trapping

Characterizing the mechanical properties of single colloids is a central problem in soft matter physics. It also plays a key role in cell biology through biopolymer condensates, which function as membraneless compartments. Such systems can also malfunction, leading to the onset of a number of diseases, including many neurodegenerative diseases; the functional and pathological condensates are commonly differentiated by their mechanical signature. Probing the mechanical properties of biopolymer condensates at the single particle level has, however, remained challenging. In this study, we demonstrate that acoustic trapping can be used to profile the mechanical properties of single condensates in a contactless manner. We find that acoustic fields exert the acoustic radiation force on condensates, leading to their migration to a trapping point where acoustic potential energy is minimized. Furthermore, our results show that the Brownian motion fluctuation of condensates in an acoustic potential well is an accurate probe for their bulk modulus. We demonstrate that this framework can detect the change in the bulk modulus of polyadenylic acid condensates in response to changes in environmental conditions. Our results show that acoustic trapping opens up a novel path to profile the mechanical properties of soft colloids at the single particle level in a non-invasive manner with applications in biology, materials science, and beyond.

biophysics↗

bioRxiv2022

Rapid protein stability prediction using deep learning representations

Predicting the thermodynamic stability of proteins is a common and widely used step in protein engineering, and when elucidating the molecular mechanisms behind evolution and disease. Here, we present RaSP, a method for making rapid and accurate predictions of changes in protein stability by leveraging deep learning representations. RaSP performs on-par with biophysics-based methods and enables saturation mutagenesis stability predictions in less than a second per residue. We use RaSP to calculate [~] 300 million stability changes for nearly all single amino acid changes in the human proteome, and examine variants observed in the human population. We find that variants that are common in the population are substantially depleted for severe destabilization, and that there are substantial differences between benign and pathogenic variants, highlighting the role of protein stability in genetic diseases. RaSP is freely available--including via a Web interface--and enables large-scale analyses of stability in experimental and predicted protein structures.

biophysics↗

bioRxiv2021

A structural biology community assessment of AlphaFold 2 applications

Most proteins fold into 3D structures that determine how they function and orchestrate the biological processes of the cell. Recent developments in computational methods have led to protein structure predictions that have reached the accuracy of experimentally determined models. While this has been independently verified, the implementation of these methods across structural biology applications remains to be tested. Here, we evaluate the use of AlphaFold 2 (AF2) predictions in the study of characteristic structural elements; the impact of missense variants; function and ligand binding site predictions; modelling of interactions; and modelling of experimental structural data. For 11 proteomes, an average of 25% additional residues can be confidently modelled when compared to homology modelling, identifying structural features rarely seen in the PDB. AF2-based predictions of protein disorder and protein complexes surpass state-of-the-art tools and AF2 models can be used across diverse applications equally well compared to experimentally determined structures, when the confidence metrics are critically considered. In summary, we find that these advances are likely to have a transformative impact in structural biology and broader life science research.

biophysics↗