Supplementary files including new predictor and benchmarks - Part of the review process of "Exo-Tox: Identifying Exotoxins from secreted bacterial proteins"

Citation: Krueger, Tanja and Jimenez Soto, Luisa F.: Supplementary files including new predictor and benchmarks - Part of the review process of "Exo-Tox: Identifying Exotoxins from secreted bacterial proteins". 8. July 2025. Open Data LMU. 10.5282/ubm/data.665

Abstract

Background: Bacterial exotoxins are secreted proteins able to affect target cells, and associated with diseases. Their accurate identification can enhance drug discovery and ensure the safety of bacteria-based medical applications. How- ever, current toxin predictors prioritize broad coverage by mixing toxins from multiple biological kingdoms and diverse control sets. This general approach has proven sub-optimal for identifying niche toxins, such as bacterial exotoxins. Recent Protein Language Models offer an opportunity to improve toxin prediction by capturing global sequence context and biochemical properties from protein sequences. Results: We introduce Exo-Tox, a specialized predictor trained exclusively on curated datasets of bacterial exotoxins and secreted non-toxic bacterial proteins,represented as embeddings by Protein Language Models. Compared to Basic Local Alignment Search Tool (BLAST)-based methods and generalized toxin predictors, Exo-Tox outperforms across multiple metrics, achieving an Matthews correlation coefficient > 0.9. Notably, Exo-Tox’s performance remains robust regardless of protein length or the presence of signal peptides. We analyze its limited transfer-ability to bacteriophage proteins and non-secreted proteins. These Datasets are part of the article published in BioData Mining : Krueger, T., Durmaz, D. & Jimenez-Soto, L. Exo-Tox: Identifying Exotoxins from secreted bacterial proteins. BioData Mining 18, 52 (2025). https://doi.org/10.1186/s13040-025-00469-2. Use the following link to access: https://rdcu.be/eAeOs

Uncontrolled Keywords

foldseek, bacterial toxins, exotoxins, embeddings, protein Language Models

References

Krueger, Tanja and Durmaz, Damla A. and Jimenez Soto, Luisa F.: Exo-Tox: Identifying Exotoxins from secreted bacterial proteins. 2025. Open Data LMU. 10.5282/ubm/data.576

• Item Type: Other
• Contact Person: Jimenez-Soto, Luisa F
• E-Mail of Contact: l.jimenez at lmu.de
• Subjects: Mathematics, Computer Science and Statistics; Biology
• Dewey Decimal Classification: 500 Natural sciences and mathematics; 500 Natural sciences and mathematics > 570 Life sciences