Gregory Brian Gloor

The paper introduces Murmurent, an agentic AI environment that supports biomedical research collaboration by providing infrastructure, specialized agents, memory, traceability, SOP enforcement, and multi-user capacity.

The paper develops PHYCUT, a CRISPR/Cas9 system for multiplex genome editing in the diatom Phaeodactylum tricornutum, demonstrating its use to reduce fucosylation of secreted proteins.

The paper improves deep learning models for predicting sgRNA/Cas9 activity by optimizing data curation and filtering, resulting in better generalization across species and data types.

This study shows that scale misspecification in differential expression analyses leads to poor false discovery rate control and provides guidance on choosing appropriate scale uncertainty.

The paper proposes a roadmap for fair reuse of public microbiome data by introducing a Data Reuse Information tag (DRI) to facilitate communication between data creators and consumers.

This study uses machine learning to predict SaCas9 activity in bacteria, revealing that PAM adenine methylation and flanking sequences significantly influence activity.

This study analyzes vaginal metatranscriptomes to identify functional subgroups and microbial strategies in bacterial vaginosis, highlighting functional differences over taxonomic ones.

The paper introduces scale models to improve microbiome and gene expression analysis by reducing false positives and negatives compared to traditional normalizations.

This paper introduces scale models to improve transcriptomic analysis by reducing false positives/negatives and enhancing transparency, using the ALDEx2 package.

The paper develops a transfer learning model, crisprHAL, to improve SpCas9/sgRNA activity prediction using small high-quality datasets.

The paper improves functional annotations of metagenomic bacteriophage by combining protein structure prediction and homology search in a naphthenic acid-degrading microbial community.

This paper presents a telomere-to-telomere genome assembly of Phaeodactylum tricornutum using Oxford Nanopore long reads, identifying 25 nuclear chromosomes and improving genetic engineering opportunities.

The paper identifies type IV conjugation systems in metagenomic data and shows they differ between age and geographic cohorts, suggesting their use as biomarkers and highlighting limitations of current binning methods.

The paper presents a strategy to complete and validate metagenome-assembled genomes using a combination of short and ultra-long reads, resulting in multiple complete genomes from a bacterial community.

This paper provides a guide for applying compositional data analysis to NGS count data, emphasizing log-ratio transformations and addressing limitations of conventional normalization methods.