Biology subjects

Blay, V.

Publications and source records attributed to Blay, V..

4 recordsLinked to original sources

bioRxiv2025

SCORCH2: a generalised heterogeneous consensus model for high-enrichment interaction-based virtual screening

The discovery of effective therapeutics remains a complex, costly, and time-consuming endeavor, characterized by high failure rates and significant resource investments. A central bottleneck in early-stage drug discovery is identifying suitable hit compounds with moderate affinity for known biological targets. Although advancements have been made, current in silico virtual screening methods are subject to limitations, including model overfitting, data bias, and constrained interpretability in their predictive processes. In this study, we present SCORCH2, a machine learning-based framework designed to enhance both the performance and interpretability of virtual screening by leveraging interaction features. Compared with its predecessor SCORCH, SCORCH2 exhibits superior predictive accuracy and generalizability across a wide range of biological targets. Importantly, SCORCH2 demonstrates robust hit identification capabilities on previously unseen targets, indicating strong transferability. These results highlight the potential of SCORCH2 as a valuable tool in accelerating drug discovery, offering reliable predictive capabilities while improving the interpretability of virtual screening models.

bioinformatics↗

bioRxiv2024

Biosynth Pipeline: An integrated platform for merging the computational design of chimeric type I polyketide synthases with enzymatic pathways for chemical biosynthesis

Synthetic biology offers the promise of manufacturing chemicals more sustainably than petrochemistry. Yet, both the rate at which biomanufacturing can synthesize these molecules and the net chemical accessible space are limited by existing pathway discovery methods which can often rely on arduous literature searches. Here, we present an automated retrobiosynthesis tool, BioPKS Pipeline, that simultaneously tackles both problems by integrating multifunctional type I polyketide synthases (PKSs) with monofunctional enzymes to propose the synthesis of desired target chemicals via two new tools: DORAnet and RetroTide. While monofunctional enzymes are valuable for carefully decorating a substrates carbon backbone, they typically cannot expand the backbone itself. PKSs can, instead, predictably do this through their unique ability to catalyze carbon-carbon bond formation reactions iteratively. We have evaluated the performance of BioPKS Pipeline against a previously published set of 155 molecules of interest for biomanufacturing, and report that BioPKS Pipeline could produce exact designs for 93 of them, as well as pipelines to a chemically similar product for most of the remaining molecules. Furthermore, BioPKS Pipeline successfully proposes biosynthetic routes for complex therapeutic natural products (cryptofolione and basidalin) for which no known biosynthetic pathway currently exists.

bioinformatics↗

bioRxiv2022

Antibody optimization enabled by artificial intelligence predictions of binding affinity and naturalness

Traditional antibody optimization approaches involve screening a small subset of the available sequence space, often resulting in drug candidates with suboptimal binding affinity, developability or immunogenicity. Based on two distinct antibodies, we demonstrate that deep contextual language models trained on high-throughput affinity data can quantitatively predict binding of unseen antibody sequence variants. These variants span a KD range of three orders of magnitude over a large mutational space. Our models reveal strong epistatic effects, which highlight the need for intelligent screening approaches. In addition, we introduce the modeling of "naturalness", a metric that scores antibody variants for similarity to natural immunoglobulins. We show that naturalness is associated with measures of drug developability and immunogenicity, and that it can be optimized alongside binding affinity using a genetic algorithm. This approach promises to accelerate and improve antibody engineering, and may increase the success rate in developing novel antibody and related drug candidates.

bioinformatics↗

bioRxiv2021

Genetic interplay between type II topoisomerase enzymes and chromosomal ccdAB toxin-antitoxin in E. coli

Fluoroquinolones are one of the most widely used class of antibiotics. They target two type II topoisomerase enzymes: gyrase and topoisomerase IV. Resistance to these drugs, which is largely caused by mutations in their target enzymes, is on the rise and becoming a serious public health risk. In this work, we analyze the sequences of 352 extraintestinal E. coli clinical isolates to gain insights into the selective pressures shaping the type II topoisomerase mutation landscape in E. coli. We identify both Quinolone Resistance-Determining Region (QRDR) and non-QRDR mutations, outline their mutation trajectories, and show that they are likely driven by different selective pressures. We confirm that ciprofloxacin resistance is specifically and strongly associated with QRDR mutations. By contrast, non-QRDR mutations are associated with the presence of the chromosomal version of ccdAB, a toxin-antitoxin operon, where the toxin CcdB is known to target gyrase. We also find that ccdAB and the evolution of QRDR mutation trajectories are partially incompatible. Finally, we identify partial deletions in CcdB and additional mutations that likely facilitate the compatibility between the presence of the ccdAB operon and QRDR mutations. These "permissive" mutations are all found in ParC (a topoisomerase IV subunit). This, and the fact that CcdB-selected mutations frequently map to topoisomerase IV, strongly suggests that this enzyme (in addition to gyrase) is likely a target for the toxin CcdB in E. coli, although an indirect effect on global supercoiling cannot be excluded. This work opens the door for the use of the presence of ccdB and of the proposed permissive mutations in the genome as genetic markers to assess the risk of quinolone resistance evolution and implies that certain strains may be genetically more refractory to evolving quinolone resistance through mutations in target enzymes.

microbiology↗