Sumeet Patiyal

The paper identifies transient isoform remodeling as a key mechanism in early MAPKi resistance in melanoma, showing it is distinct from gene expression changes and linked to splicing regulators and developmental pathways.

This paper develops HEiST, a deep learning method to infer spatial transcriptomes from H&E slides, identifies two prognostic Spatiotypes in HNSC, and shows they predict treatment response better than existing biomarkers.

Path2Space predicts spatial gene expression and tumor microenvironment from histopathology slides, identifying new breast cancer subgroups and improving biomarker prediction.

This paper develops a machine learning framework, EXPRESSO, to predict treatment response from tumor transcriptomics across multiple cancer types and therapies, showing improved performance over existing methods.

This paper introduces TIME_ACT, a transcriptomic signature predicting immune checkpoint blockade response across cancers, validated using histopathology and multiple cohorts.

PEPlife2 is an updated manually curated database with 4,412 peptide half-life entries, including structural data and advanced analysis tools.

This paper develops a machine learning model and software tool to predict NF-κB inhibitors from chemical libraries, achieving reliable predictions validated against experimental data.

The paper develops a hybrid method to predict IFN-gamma inducing peptides in human and mouse, achieving higher AUROC scores than existing methods.

This paper introduces HLA-DR4Pred2, an improved method for predicting HLA-DRB1*04:01 binders using a larger dataset and hybrid approaches, achieving higher AUC scores than existing methods.

This paper develops a machine learning model to predict antifreeze proteins using evolutionary information, achieving higher accuracy than existing methods on an independent dataset.

The paper performs a meta-analysis of 8.6 million antimicrobial peptides to identify class-specific therapeutic candidates using machine learning and structural analysis.

This study develops a hybrid method combining alignment-based and alignment-free approaches to predict transcription factors with high accuracy, achieving an AUC of 0.99 on an independent dataset.

This paper develops a deep learning model using single-cell transcriptomics to classify HNSCC and distinguish HPV+ from HPV- patients, achieving high AUROC scores on a validation dataset.

The paper develops a hybrid model for predicting both linear and conformational B-cell epitopes using machine learning and BLAST, achieving an AUC of 0.83 on an independent dataset.

The paper develops a logistic regression model with a propensity index panel of ten genes to accurately screen prostate cancer, achieving 96.36% accuracy and 0.99 AUROC.

This paper develops a hybrid model combining motif-based and machine learning approaches to predict exosomal proteins with high accuracy, achieving AUROC of 0.85 on an independent dataset.

This study develops a deep learning model to predict Alzheimer's disease using single-cell transcriptomics data, achieving an AUC of 0.75-0.84 on validation datasets.

Hmrbase2 is an updated database containing hormone and receptor information from 803 organisms, with improved data content and search features.

This paper develops computational tools to predict celiac disease-associated epitopes and motifs in proteins, achieving high accuracy using machine learning and motif-based approaches.

This paper develops a fast method for predicting multi-label subcellular localization of mRNA at genome scale, achieving high AUROC scores using hybrid techniques.

The paper introduces Pprint2, a method using convolutional neural networks and evolutionary profiles to predict RNA-interacting residues in proteins, achieving an AUC of 0.82.

This paper develops in-silico models to predict infectious Influenza A virus strains from genome and protein sequences, achieving high AUC scores and providing a web tool for pandemic preparedness.

The paper introduces a software package that computes over 50,000 features for protein/peptide function prediction, including residue-level annotations.

DBpred predicts DNA-interacting residues in proteins using deep learning, achieving improved AUROC scores compared to existing methods.

This paper develops models to predict, scan, and design TNF-α inducing epitopes for human and mouse, achieving AUROC values up to 0.83 for human and 0.77 for mouse.

This paper develops methods to predict and design T1DM-associated peptides, achieving high accuracy with a hybrid approach and a webserver for peptide prediction.

This paper develops an ensemble model to predict phage-host interactions for phage therapy, achieving higher accuracy than existing methods across multiple taxonomic levels.

The paper develops HLAncPred, an AI-based method to predict non-classical HLA (HLA-G and HLA-E) binders, achieving high AUC scores and providing a web server for this purpose.

This paper introduces IL-13Pred, a method to predict IL-13 inducing peptides in proteins, with high accuracy for managing COVID-19 severity.

The paper benchmarks mutation calling techniques for liver cancer prognosis using VCF and MAF files, finding MuTect2's VCF files best predict high-risk patients with high AUROC and HR.