Biology subjects

Gauthier, N. P.

Publications and source records attributed to Gauthier, N. P..

7 recordsLinked to original sources

bioRxiv2023

Deep learning prediction of enzyme optimum pH

The relationship between pH and enzyme catalytic activity, especially the optimal pH (pHopt) at which enzymes function, is critical for biotechnological applications. Hence, computational methods to predict pHopt will enhance enzyme discovery and design by facilitating accurate identification of enzymes that function optimally at specific pH levels, and by elucidating sequence-function relationships. In this study, we proposed and evaluated various machine-learning methods for predicting pHopt, conducting extensive hyperparameter optimization, and training over 11,000 model instances. Our results demonstrate that models utilizing language model embeddings markedly outperform other methods in predicting pHopt. We present EpHod, the best-performing model, to predict pHopt, making it publicly available to researchers. From sequence data, EpHod directly learns structural and biophysical features that relate to pHopt, including proximity of residues to the catalytic center and the accessibility of solvent molecules. Overall, EpHod presents a promising advancement in pHopt prediction and will potentially speed up the development of enzyme technologies.

biochemistry↗

bioRxiv2023

Simultaneous enhancement of multiple functional properties using evolution-informed protein design

Designing optimized proteins is important for a range of practical applications. Protein design is a rapidly developing field that would benefit from approaches that enable many changes in the amino acid primary sequence, rather than a small number of mutations, while maintaining structure and enhancing function. Homologous protein sequences contain extensive information about various protein properties and activities that have emerged over billions of years of evolution. Evolutionary models of sequence co-variation, derived from a set of homologous sequences, have proven effective in a range of applications including structure determination and mutation effect prediction. In this work we apply one of these models (EVcouplings) to computationally design highly divergent variants of the model protein TEM-1 {beta}-lactamase, and characterize these designs experimentally using multiple biochemical and biophysical assays. Nearly all designed variants were functional, including one with 84 mutations from the nearest natural homolog. Surprisingly, all functional designs had large increases in thermostability and most had a broadening of available substrates. These property enhancements occurred while maintaining a nearly identical structure to the wild type enzyme. Collectively, this work demonstrates that evolutionary models of sequence co-variation (1) are able to capture complex epistatic interactions that successfully guide large sequence departures from natural contexts, and (2) can be applied to generate functional diversity useful for many applications in protein design.

synthetic biology↗

bioRxiv2021

BugSplit: highly accurate taxonomic binning of metagenomic assemblies enables genome-resolved metagenomics

A large gap remains between sequencing a microbial community and characterizing all of the organisms inside of it. Here we develop a novel method to taxonomically bin metagenomic assemblies through alignment of contigs against a reference database. We show that this workflow, BugSplit, bins metagenome-assembled contigs to species with a 33% absolute improvement in F1-score when compared to alternative tools. We perform nanopore mNGS on patients with COVID-19, and using a reference database predating COVID-19, demonstrate that BugSplits taxonomic binning enables sensitive and specific detection of a novel coronavirus not possible with other approaches. When applied to nanopore mNGS data from cases of Klebsiella pneumoniae and Neisseria gonorrhoeae infection, BugSplits taxonomic binning accurately separates pathogen sequences from those of the host and microbiota, and unlocks the possibility of sequence typing, in silico serotyping, and antimicrobial resistance prediction of each organism within a sample. BugSplit is available at https://bugseq.com/academic.

bioinformatics↗

bioRxiv2019

Protein structure from experimental evolution

Natural evolution encodes rich information about the structure and function of biomolecules in the genetic record. Previously, statistical analysis of co-variation patterns in natural protein families has enabled the accurate computation of 3D structures. Here, we explored whether similar information can be generated by laboratory evolution, starting from a single gene and performing multiple cycles of mutagenesis and functional selection. We evolved two bacterial antibiotic resistance proteins, {beta}-lactamase PSE1 and acetyltransferase AAC6, and obtained hundreds of thousands of diverse functional sequences. Using evolutionary coupling analysis, we inferred residue interactions in good agreement with contacts in the crystal structures, confirming genetic encoding of structural constraints in the selected sequences. Computational protein folding with contact constraints yielded 3D structures with the same fold as that of natural relatives. Evolution experiments combined with inference of residue interactions from sequence information opens the door to a new experimental method for the determination of protein structures.

evolutionary biology↗

bioRxiv2019

Perturbation biology links temporal protein changes to drug responses in a melanoma cell line

Data-driven mathematical modeling of biological systems has enormous potential to understand and predict the interplay between molecular and phenotypic response to perturbation, and provides a rational approach to the nomination of therapies in complex diseases such as cancer. Melanoma is a particularly debilitating disease for which most therapies eventually fail as resistance to chemotherapy and targeted drugs develop. We have previously applied an iterative experimental-computational modeling approach, termed perturbation biology, to predict and test effective drug combinations in melanoma cell lines [1].\n\nIn this work, we extend our analysis framework to derive models of temporally-acquired perturbation data that do not require prior knowledge and explicit specification of the targets of individual drugs. Specifically, we characterize the response of the melanoma cell line A2058 to 54 cancer drug combinations at 8 logarithmically spaced time points from 10 minutes to 67 hours. At each time point, 124 antibodies of proteins and phospho-proteins with broad coverage of cancer-related pathways and two phenotypes (cell number and apoptosis) were measured. These data are used to infer interactions in ordinary differential equation-based models that capture temporal aspects of the drug perturbation data. This network representation of drug-protein, protein-protein, and protein-phenotype interactions can be used to identify new logical (not necessarily direct biochemical) interactions. The agreement between the predicted phenotypic response and corresponding data for unseen drug perturbations has a Pearsons correlation coefficient of 0.79. We further use model predictions to nominate effective combination therapies and perform experimental validation of the highest ranked combinations.\n\nThis new data-driven modeling framework is a step forward in perturbation biology as it incorporates the temporal aspect of data. This work therefore opens the door to a new understanding of dynamic drug responses at a molecular level.

systems biology↗

bioRxiv2018

3D protein structure from genetic epistasis experiments

High-throughput experimental techniques have made possible the systematic sampling of the single mutation landscape for many proteins, defined as the change in protein fitness as the result of point mutation sequence changes. In a more limited number of cases, and for small proteins only, we also have nearly full coverage of all possible double mutants. By comparing the phenotypic effect of two simultaneous mutations with that of the individual amino acid changes, we can evaluate epistatic effects that reflect non-additive cooperative processes. The observation that epistatic residue pairs often are in contact in the 3D structure led to the hypothesis that a systematic epistatic screen contains sufficient information to identify the 3D fold of a protein. To test this hypothesis, we examined experimental double mutants for evidence of epistasis and identified residue contacts at 86% accuracy, including secondary structure elements and evidence for an alternative all--helical conformation. Positively epistatic contacts - corresponding to compensatory mutations, restoring fitness - were the most informative. Folded models generated from top-ranked epistatic pairs, when compared with the known structure, were accurate within 2.4 [A] over 53 residues, indicating the possibility that 3D protein folds can be determined experimentally with good accuracy from functional assays of mutant libraries, at least for small proteins. These results suggest a new experimental approach for determining protein structure.

systems biology↗

bioRxiv2017

Protein Profiling In Cancer Cell Lines And Tumor Tissue Using Reverse Phase Protein Arrays

Reverse phase protein array (RPPA) technology is an antibody-based high-throughput assay for protein profiling of biological specimens that allows for many measurements with very small amounts of cell lysate. Here, we report the sensitivity, reproducibility, and accuracy of a particular RPPA platform called Zeptosens. We customized the RPPA protocol for our in-house setup, and measured more than 80 total protein and phospho-protein levels in various cancer samples, including cell lines, organoids, tumor chunks, core needle biopsies, and laser-capture microdissected tissue samples. We discuss pros and cons of the RPPA platform, and describe results from profiling 15 cancer cell line cells using RPPA.

systems biology↗