Selected Work

Open-source tools and peer-reviewed research contributions in bioinformatics, metagenomics, and carbohydrate-active enzyme analysis.

The work spans population-scale cohort screening, RNA-seq differential expression, phylogenetic placement, scientific software development, and HPC pipeline engineering — methods that transfer across biological systems. If the problem looks familiar, a scoping call will quickly establish whether there's a fit.

Open Source Tools

SACCHARIS 2

Sequence Analysis and Clustering of CarboHydrate Active enzymes for Rapid Informed prediction of Specificity

View on GitHub

A bioinformatics pipeline for phylogenetic functional prediction of carbohydrate-active enzymes (CAZymes). Embeds novel sequences within phylogenetic trees of biochemically characterised CAZymes to enable informed predictions of substrate specificity.

Methods in Molecular Biology, Springer Nature, 2024
PythonBiocondaCLIGUI BLAST+DIAMONDHMMERMUSCLEModelTest-NGFastTreeRAxML

Role: Lead developer who rewrote the original Perl pipeline in Python, added proper packaging and Bioconda distribution, and designed and implemented a full GUI from scratch.

Publications

Bespoke plant glycoconjugates for gut microbiota-mediated drug targeting

A targeted drug delivery system (GlycoCaging) using plant oligosaccharides to trigger site-specific release of anti-inflammatory drugs via gut bacterial glycosidases, validated in inflammatory bowel disease models.

MetagenomicsHPCPython

Alex's contribution

Owned the full metagenomics component — from raw HMP2 reads to the figures underpinning the paper's central targeting claim.

Designed and executed the metagenomics component: aligned HMP2 longitudinal fecal metagenomes (n=124 across healthy controls, ulcerative colitis, and Crohn's Disease cohorts) against xyloglucan utilization locus references using Magic-BLAST on the Digital Research Alliance of Canada HPC cluster. Longitudinal presence/absence tracking of five GH5_4 endo-xyloglucanase genes across up to 24 bi-weekly timepoints per individual, with dual-threshold validation and systematic QC annotation. Downstream analysis and visualization in Python (pandas, NumPy, seaborn, Matplotlib).

Transcriptional delineation of polysaccharide utilization loci in Segatella copri DSM18205 and co-culture with exemplar Bacteroides species on dietary plant glycans

RNA-seq characterisation of polysaccharide utilization loci (PULs) and CAZymes in a key human gut commensal, revealing differential transcriptional responses across plant polysaccharides including mixed-linkage β-glucans, xyloglucans, xylans, pectins, and inulin.

RNA-seqDESeq2Visualization

Alex's contribution

Ran the RNA-seq analysis pipeline and produced the RNA-seq data visualisations for the published manuscript.

Analyzed RNA-seq data and generated all data visualisations. Pipeline: FastQC/MultiQC for QC, Trimmomatic for trimming, Salmon for transcript quantification, and DESeq2 for differential expression analysis.

The molecular basis of cereal mixed-linkage β-glucan utilization by the human gut bacterium Segatella copri

Structural and mechanistic study of how S. copri recognises and degrades cereal β-glucans, combining biochemistry, X-ray crystallography, and molecular analysis to characterise a novel MLG-specific PUL.

MetagenomicsPhylogeneticsSACCHARIS

Alex's contribution

Conducted population-scale metagenomics screening across four continents and GH5_4 phylogenetic analysis contextualising the paper's enzyme specificity findings.

Conducted population-scale metagenomics screening of human stool metagenomes (NCBI + HMP2 datasets representing cross-sections from North America, Europe, Asia, and Africa) for presence of Segatella MLG-PULs across 15 species. Also performed GH5_4 phylogenetic analysis using SACCHARIS.

Interested in working together?

Seen something here that maps to a problem in your lab? A 30-minute scoping call, no charge, is the fastest way to find out.