Colby T. Ford

Hoike generates condition-specific bulk transcriptomes using a joint-embedding predictive architecture and diffusion models, conditioned on normal tissue profiles.

Kukulu reconstructs antibody CDR loops using a diffusion model with a structure-aware architecture, improving loop design by integrating sequence-structure constraints.

This study shows that short-term ex vivo culture of Plasmodium vivax enriches late asexual stages and selectively filters clones without inducing major transcriptional changes.

This study analyzes the spread and adaptation of H5N1 influenza in North America, finding increased host range and improved mammalian replication, with reduced antiviral effectiveness in some isolates.

This paper introduces peleke-1, a suite of protein language models fine-tuned for generating targeted antibody sequences from antigen inputs.

This study computationally models H5 influenza variants against HA1-neutralizing antibodies, showing weakening antibody binding over time and increased immune evasion risks.

This paper provides a comprehensive guide for quality control metrics and mitigation strategies for 11 epigenetics and transcriptomics assays to improve exposure signature discovery.

This study uses AI to design novel antibodies against recent H5N1 influenza strains, showing they bind similarly to known therapeutic antibodies.

The paper uses virtual screens to predict cytokine-nucleocapsid protein interactions across coronaviruses, comparing AlphaFold2-Multimer and HADDOCK, and introduces a new algorithm for assessing binding interfaces.

This study investigates mecp2 function in zebrafish by combining transcriptomic and behavioral analyses, revealing conserved transcriptional regulation and novel roles in visual stimulus responses.

This study predicts binding affinity between SARS-CoV-2 variants BA.2.86 and JN.1 and antibodies/ACE2 using in silico modeling, finding minor reductions in affinity but no significant immune escape.

The paper analyzes Plasmodium falciparum CSP variants' interactions with human immunoproteins to explain RTS,S vaccine efficacy in Ghana, finding that peptide length affects binding affinity and immune response.

This study predicts structural changes in the SARS-CoV-2 Omicron variant's spike protein receptor-binding domain and their impact on neutralizing antibody interactions.

This study analyzes whole genome sequences of Plasmodium vivax isolates from Ethiopia to identify sequence and structural polymorphisms in erythrocyte binding genes, revealing positive selection and gene copy number variations associated with host-parasite interactions.

This paper develops ensemble machine learning models to predict Artemisinin resistance and parasite clearance rates in Plasmodium falciparum isolates using transformed data and preprocessing techniques.

StrainHub constructs pathogen transmission networks using phylogenetic trees and metadata to identify key nodes for infection control.

The paper predicts the effect of NAGLU missense mutations on enzymatic activity using machine learning and PolyPhen-2 scores, finding they are good predictors of functionality.