Supplementary code data files for the thesis "A Population-Based Approach to Genomic Island Analysis"
Bibliographic record
Abstract
Bacteria are rich with functional diversity enhanced by their ability to readily adopt and transmit novel traits via horizontal gene transfer. Genomic islands (GIs), defined as clusters of genes of probable horizontal origin integrated into the chromosome, are of particular interest because they disproportionately encode novel features, including antimicrobial resistance genes and virulence factors. As genomic approaches to investigating pathogen outbreaks and characterizing the dissemination of genes across microbial populations advance, there is a need for specialized tools to characterize GIs in population datasets. However, previous tools only allowed for GI prediction in single genomes. Additionally, it was unclear how well GIs are predicted in incomplete genomes. The aim of my thesis was to facilitate reliable GI comparison in multi-genome datasets and investigate GIs involved in disseminating genes of interest within datasets of clinical relevance. IslandCompare is a newly-released, web-based platform designed to handle hundreds of genomes. My work focused on incorporating novel functionality for clustering GIs and ensuring consistent cross-genome predictions. This, coupled with the facilitated visualization platform, allows users to rapidly identify differences in GI content across genomes. I also led the establishment of a centralized database of curated GIs, which already includes entries from key pathogenic species. Targeted predictions of these curated GI sequences are being incorporated into IslandCompare alongside functional information. I developed an approach for predicting curated Salmonella enterica GIs. My evaluation of draft genomes revealed that GI predictions are often missed by current sequence composition-based methods, especially when interrupted by contig breaks. Predictions are also less sensitive in metagenome-assembled genomes. An investigation of Enterococcus faecium genomes from disparate habitats and continents revealed that habitat plays a greater role than geography in phylogenetic differentiation and associated accessory gene contents. Analysis with IslandCompare led to the identification of GI clusters involved in the dissemination of antimicrobial resistance genes and virulence factors, including vancomycin resistance and surface adhesion genes. My collective work will facilitate a better understanding of computational GI prediction and comparison of GI contents across population datasets, including for cases where GIs play a role in the dissemination of genes of clinical and environmental importance.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.026 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.003 | 0.006 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.004 | 0.003 |
| Open science | 0.004 | 0.004 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.739 | 0.346 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".