CSM-Toxin: A Web-Server for Predicting Protein Toxicity

Abstract

Biologics are one of the most rapidly expanding classes of therapeutics, but can be associated with a range of toxic properties. In small-molecule drug development, early identification of potential toxicity led to a significant reduction in clinical trial failures, however we currently lack robust qualitative rules or predictive tools for peptide- and protein-based biologics. To address this, we have manually curated the largest set of high-quality experimental data on peptide and protein toxicities, and developed CSM-Toxin, a novel in-silico protein toxicity classifier, which relies solely on the protein primary sequence. Our approach encodes the protein sequence information using a deep learning natural languages model to understand “biological” language, where residues are treated as words and protein sequences as sentences. The CSM-Toxin was able to accurately identify peptides and proteins with potential toxicity, achieving an MCC of up to 0.66 across both cross-validation and multiple non-redundant blind tests, outperforming other methods and highlighting the robust and generalisable performance of our model. We strongly believe the CSM-Toxin will serve as a valuable platform to minimise potential toxicity in the biologic development pipeline. Our method is freely available as an easy-to-use webserver.

Towards a Dataset for State of the Art Protein Toxin Classification

“…Contemporary toxin classification methods that make use of large language models are UniDL4BioPep (30) and CSM-Toxin (31). UniDL4BioPep uses pre-trained protein language model embeddings as features into additional deep learning layers.…”

Designing Combinatorial Multiepitope Vaccine Candidates against Scrub Typus by a Proteome-wide Immunoinfomatics Approach.

“…Allergenicity of the antigens were predicted by AlgPred2.0 (RRID:SCR_018780) 29 with a threshold of 0.3 for AAC based RF for allergenicity, and AllerCatPro2.0 30 . Toxicity of the MEVs were predicted through CSM-Toxin 31 using default parameters.…”

Engineering G protein‐coupled receptors for stabilization

“…The atom pharmacophores are characteristics belonging to eight possible classes: hydrophobic, positive, negative, hydrogen acceptor, hydrogen donor, aromatic, sulfur, and neutral. These signatures have shown to be an effective and efficient method to model protein residue environment, its geometry and physicochemical properties, information that has been used to predict the effects of mutations on protein stability and affinity to its partners (Myung, Pires, & Ascher, 2020 ; Myung, Rodrigues, et al, 2020 ; Nguyen et al, 2021 ; Pires et al, 2014 , 2016 ; Pires & Ascher, 2016 , 2017 ; Rodrigues et al, 2019 , 2021a ; 2021b , 2024 ; Rodrigues & Ascher, 2022 , 2023 ; Ryu et al, 2023 ), pharmacodynamic and pharmacokinetics (Al‐Jarf et al, 2021 ; de Sa et al, 2022 ; Iftkhar et al, 2022 ; Morozov et al, 2023 ; Pires et al, 2015 , 2022 ; Pires & Ascher, 2020 ; Rodrigues et al, 2021c , 2022 ; Velloso et al, 2021 ), and identify drug resistance (Hawkey et al, 2018 ; Karmakar et al, 2018 , 2019 , 2020 ; Portelli et al, 2020 ; Portelli, Heaton, & Ascher, 2023 ; Zhan et al, 2021 ; Zhou et al, 2021 ) and disease mutations (Jessen‐Howard et al, 2023 ; Karmakar et al, 2022 ; Lai et al, 2021 ; Portelli et al, 2021 ; Portelli, Albanaz, et al, 2023 ).…”