Toxicity emerges as a prominent challenge in the design of therapeutic peptides, causing the failure of numerous peptides during clinical trials. In 2013, our group developed ToxinPred, a computational method that has been extensively adopted by the scientific community for predicting peptide toxicity. In this paper, we propose a refined variant of ToxinPred that showcases improved reliability and accuracy in predicting peptide toxicity. Initially, we used BLAST for alignment-based toxicity prediction, yet coverage was limited. We adopted a motif-based approach with MERCI software to identify unique toxic patterns. Despite specificity gains, sensitivity was compromised. We developed alignment-free methods using machine/deep learning, achieving a balance sensitivity and specificity of prediction. A deep learning model (ANN – LSTM with fixed sequence length) developed using one-hot encoding attained a 0.93 AUROC and 0.71 MCC on independent data. The machine learning model (extra tree) developed using compositional features of peptides achieved 0.95 AUROC and 0.78 MCC. Lastly, we developed hybrid or ensemble methods combining two or more models to enhance performance. Hybrid approaches, including motif-based and machine learning, achieved a 0.98 AUROC and 0.81 MCC. Evaluation on independent data demonstrated our method’s superiority. To cater to the needs of the scientific community, we have developed a standalone software, pip package and web-based server ToxinPred3 ( https://github.com/raghavagps/toxinpred3 and https://webs.iiitd.edu.in/raghava/toxinpred3/ ) . Author’s Biography: Anand Singh Rathore is currently pursuing a Ph.D. in Computational Biology at the Department of Computational Biology, Indraprastha Institute of Information Technology, New Delhi, India. Akanksha Arora is currently pursuing a Ph.D. in Computational Biology at the Department of Computational Biology, Indraprastha Institute of Information Technology, New Delhi, India. Shubham Choudhury is currently pursuing a Ph.D. in Computational Biology at the Department of Computational Biology, Indraprastha Institute of Information Technology, New Delhi, India. Purava Tijare is a Project Fellow in Computational Biology at the Department of Computational Biology, Indraprastha Institute of Information Technology, New Delhi, India. Gajendra P. S. Raghava is currently working as a Professor and Head of the Department of Computational Biology, Indraprastha Institute of Information Technology, New Delhi, India. Highlights: Implementation of alignment or similarly based techniques for predicting toxic peptides. Discovery of toxicity-associated patterns and identification of toxic regions in peptides. Development of machine and deep learning-based models for toxicity prediction. Ensemble methods that combine alignment-based and alignment-free methods. Web server and standalone software package for screening toxicity in peptides/proteins.

Several formats, including FASTA, PIR, GenBank, EMBL, and GCG, have been developed for representing protein sequences composed of natural amino acids. Among these, FASTA remains the most widely used due to its simplicity and human readability. However, FASTA lacks the capability to represent chemically modified or non-natural residues, as well as structural annotations and mutations in protein variants. To address some of these limitations, the PEFF format was recently introduced as an extension of FASTA. Additionally, formats such as HELM and BILN have been proposed to represent amino acids and their modifications at the atomic level. Despite their advancements, these formats have not achieved widespread adoption within the bioinformatics community due to their complexity. To complement existing formats and overcome current challenges, we propose a new format called MAP (Modification and Annotation in Proteins), which enables comprehensive annotation of protein sequences. MAP introduces meta tags in the header for protein-level annotations and inline tags within the sequence for residue-level modifications. In this format, standard one-letter amino acid codes are augmented with curly-brace tags to denote various modifications, including phosphorylation, acetylation, non-natural residues, cyclization, and other residue-specific features. The header metadata also captures information such as organism, function, and sequence variants. We describe the structure, objectives, and capabilities of the MAP format and demonstrate its application in bioinformatics, particularly in the domain of protein therapeutics. To facilitate community adoption, we are developing a comprehensive suite of MAP-format resources, including a detailed manual, annotated datasets, and conversion tools, available at http://webs.iiitd.edu.in/raghava/maprepo/.

Several peptide-based drugs fail in clinical trials due to their toxicity or hemolytic activity against red blood cells (RBCs). Existing methods predict hemolytic peptides but not the concentration (HC50) required to lyse 50% of RBCs. In this study, we developed a classification model and regression model to identify and quantify the hemolytic activity of peptides. Our models were trained and validated on 1924 peptides with experimentally determined HC50 against mammalian RBCs. Analysis indicates that hydrophobic and positively charged residues were associated with higher hemolytic activity. Our classification models achieved a maximum AUC of 0.909 using a hybrid model of ESM-2 and a motif-based approach. Regression models using compositional features achieved R of 0.739 with R² of 0.543. Our models outperform existing methods and are implemented in the web-based platform HemoPI2 and standalone software for designing hemolytic peptides with desired HC50 values (http://webs.iiitd.edu.in/raghava/hemopi2/). Highlights Developed classification and regression models to predict hemolytic activity and HC50 values of peptides. A hybrid model combining machine learning and motif prediction excels in accuracy. Benchmarking of the existing classification methods on independent datasets. Web server, standalone software, and pip package for hemolytic activity prediction of peptides/proteins.

Most of the existing methods including our NTxPred, use a single model to predict both neurotoxic peptides and neurotoxic proteins (neurotoxins). In this study, we developed distinct models for predicting neurotoxic peptides and neurotoxins. Our peptide dataset consists of 877 neurotoxic and 877 non-neurotoxic peptides, while our protein dataset includes 775 neurotoxins and 775 non-neurotoxins. Preliminary analysis reveals that certain residues, such as cysteine, are more prevalent in both neurotoxic peptides and proteins, though their abundance differs in magnitude. First, we developed machine learning models using composition and binary profiles, achieving a maximum AUC of 0.97 for peptides and 0.85 for proteins. The performance for proteins improved from an AUC of 0.85 to 0.89 when evolutionary information was incorporated. Next, we built machine learning models using embeddings from protein language models, attaining an AUC of 0.96 for peptides and 0.94 for proteins. We also developed protein language models and achieved an AUC of 0.98 for peptides using esm2-t30 and 0.91 for proteins using esm2-t6. All models were trained and tested using 5-fold cross-validation, and final models were evaluated on an independent dataset not used in training. We further assessed protein models on the peptide dataset and vice versa, highlighting the necessity of separate models. The proposed models outperform existing methods on independent datasets. Our neurotoxicity prediction models will aid in the safety assessment of genetically modified foods and therapeutic proteins by minimizing the need for animal testing. To support the scientific community, we developed a standalone software and web server NTxPred2, for predicting and scanning neurotoxins ( https://webs.iiitd.edu.in/raghava/ntxpred2/ , https://github.com/raghavagps/ntxpred2/ ). Highlights: Development of models for predicting neurotoxic peptides and neurotoxins ML Models Using LLM Embeddings for Neurotoxin Prediction Protein language models esm2-t30 for predicting neurotoxic peptides Important for risk assessment of genetically modified foods/proteins NTxPred2 is an updated version of NTxPred for predicting neurotoxicity