Timothy P. Jenkins

This paper presents a rapid, automation-friendly pipeline for characterizing de novo protein binders without cell culture or purification, achieving binding affinity and kinetics in three days.

This paper introduces Proton-PottsMPNN, a model that designs pH-sensitive binders by explicitly modeling protonation states, resulting in pH-dependent PD-L1 binders with transition pH values between 4.0 and 5.8.

Odin-Multi is a binder design framework that optimizes sequences against multiple targets simultaneously, improving cross-reactivity and specificity compared to single-target methods.

This paper introduces InstaNovo-FM, a self-supervised foundation model for proteomics that learns from tandem mass spectra to enable transferable applications across datasets and instruments.

This study designs de novo single-domain antibodies against cobra neurotoxin and shows they protect mice from lethal venom toxicity.

This paper introduces a hybrid quantum-classical pipeline for designing MHC-binding peptides, showing that quantum-derived priors improve the yield of strong binders, especially for understudied HLA alleles.

InstaNovo-P improves phosphorylated peptide detection and site localization in phosphoproteomics by using a transformer-based model fine-tuned on phosphoproteomics data.

The paper designs a cyclic MC4R peptide agonist using deep learning and validates its efficacy in reducing food intake in mice.

This paper introduces a workflow for direct protein sequencing using AI-driven methods, validated across diverse proteins to improve accuracy and reduce ambiguity in therapeutic and scientific applications.

This study shows that mRNA-delivered consensus allergens induce neutralizing IgG responses against multiple food and pollen allergens, offering a safer and broader allergy immunotherapy approach.

This paper develops a pipeline combining yeast screening and AlphaFold3 to accelerate discovery of multi-target antibodies, finding that structural predictions improve binding specificity and solubility.

Conditioning RFdiffusion on edge beta-strand pairing yields specific binders to polar targets, with in silico success above 9.2% versus 0.98% for hotspot-only conditioning.

The paper designs protein-based inhibitors to efficiently inhibit and remove chymosin from whey, enabling sustainable cheese production and enzyme control.

This study designs a protein that precisely controls Gremlin-1 antagonism by releasing BMPs, revealing the role of interface rigidity in antagonist function.

This study benchmarks mass spectrometry-based antibody sequencing methods, revealing biases related to sequence, concentration, protease, and bioinformatics tools.

The paper analyzes 3,766 binders to identify predictive features for experimental success in de novo binder design, showing that AF3-derived metrics improve prediction accuracy.

This paper designs de novo pMHC binders using generative models to create T cell-based cancer therapeutics with high specificity and minimal cross-reactivity.

The paper assesses genetic diversity of soil-transmitted helminths across regions and evaluates its impact on molecular diagnostic targets using low-coverage genome sequencing and in vitro assays.

InstaNovo is a transformer model for de novo peptide sequencing that outperforms existing methods, enabling accurate identification of peptides and discovery of novel sequences and organisms.

ProteusAI is an open-source platform that streamlines ML-guided protein design and engineering, improving proteins and enzymes efficiently within the DBTL cycle.

This study evaluates AF2's ability to predict protein structural ensembles by comparing its results with MD simulations and NMR data for four test proteins.

This study compares three protein structure prediction tools on snake venom toxins with no reference templates, finding AlphaFold2 performs best, while ColabFold is computationally less intensive.

The paper introduces a simplified method for amplifying plasmid DNA libraries using rolling circle amplification and endonuclease treatment to improve transformation efficiency in E. coli and S. cerevisiae.

ALOHA is an AI-guided tool that quantifies venom-induced haemorrhage in mice using smartphone imaging, reducing analysis time and human error.