CPE Genomics in Hospital Reservoirs

Explain how the major carbapenemase families differ in mechanism, genetic context, and epidemiological significance.

Identify the roles of insertion sequences, transposons, integrons, and plasmids in carbapenemase-gene mobilisation.

Explain why sinks, drains, toilets, and showers can select for and sustain CPE populations and resistance plasmids.

Distinguish clonal strain transmission from horizontal transfer of a resistance plasmid.

Formulate testable genomic hypotheses linking damp-environment reservoirs, colonisation, and patient infection.

Explain command-line syntax, options, positional arguments, help pages, exit codes, and standard input and output.

Navigate directories and inspect, copy, move, compress, and checksum genomic data files safely from the shell.

Use pipes, redirection, grep, cut, sort, awk, and sed to inspect and transform genomics text files.

Create reproducible genomic software environments with Conda or Mamba and export their specifications.

Run a genomics application in Docker or Apptainer and explain how container isolation supports reproducibility.

Write a defensive Bash loop that applies one command to a sample manifest without overwriting inputs.

Capture commands, parameters, software versions, logs, and checksums for a genomic analysis step.

Organise raw data, intermediate files, results, software environments, and provenance in a reproducible project structure.

Verify file integrity, isolate identity, and short-read pairing before analysis.

Interpret per-base quality, adapter content, duplication, read length, and GC profiles in Illumina reads.

Run FastQC and MultiQC from the command line and trace each report metric back to the underlying short-read data.

Apply explicit rules for trimming, filtering, or retaining short reads based on quality-control evidence.

Run fastp with documented parameters and verify the effects of short-read adapter and quality filtering.

Use NanoPlot and an estimated genome size to interpret long-read quality and calculate nominal sequencing coverage.

Detect likely contamination or mixed-isolate sequencing using read-level taxonomic and k-mer evidence.

Record auditable pass, caution, or fail decisions for every read set.

Select short-read or hybrid assembly strategies according to the data available for each isolate.

Run Shovill or SPAdes to generate a bacterial short-read assembly and explain the de Bruijn graph model behind it.

Generate multiple complete long-read assemblies with different assemblers and read subsets for use as Autocycler inputs.

Run the Autocycler consensus workflow and explain how compression, clustering, trimming, resolution, and combination produce a consensus assembly.

Inspect Autocycler cluster metrics and graphs to decide whether automatic resolution is trustworthy or manual review is required.

Polish an Autocycler consensus assembly with Polypolish and verify the short-read-supported corrections.

Run QUAST and CheckM2 and apply an assembly-acceptance rubric to the resulting quality metrics.

Assign Enterobacter and Citrobacter species using genome-wide similarity to suitable reference genomes.

Run GTDB-Tk and skani or FastANI to resolve Enterobacter and Citrobacter species assignments and explain their reference-comparison logic.

Resolve disagreements among submitted species labels, genome-based taxonomy, and contamination evidence.

Run Bakta to annotate bacterial assemblies and interpret its principal tabular, sequence, and graphical output files.

Run command-line MLST typing and verify that the selected scheme is appropriate for the assigned species.

Run AMRFinderPlus or ResFinder with documented thresholds and interpret exact, partial, and uncertain resistance-gene matches.

Confirm carbapenemase gene alleles by inspecting read coverage and local assembly quality.

Classify carbapenemase genes as plasmid-borne, chromosomally integrated, or unresolved using convergent genomic evidence.

Choose among core-SNP, core-genome, and gene-content approaches for a defined relatedness question.

Build a core-genome alignment with Panaroo or a reference-based core-SNP alignment with Snippy and explain the assumptions of the selected approach.

Run Gubbins to detect recombinant regions and explain how recombination masking changes the clonal alignment.

Run IQ-TREE to infer a maximum-likelihood phylogeny and interpret model selection, branch support, and warning messages.

Run snp-dists on an appropriate alignment and validate the resulting pairwise-distance matrix.

Evaluate genomic-distance distributions before choosing operational cluster criteria.

Define candidate isolate clusters by combining genomic relatedness with epidemiological context rather than relying on a universal cutoff.

Encode the nested spatial relationships among hospitals, wards, rooms, damp outlets, and patient locations.

Use R and ggplot2 commands to visualise sampling effort across hospitals, dates, niches, and species.

Use ggtree commands to annotate isolate phylogenies with hospital, niche, date, and clinical-source metadata.

Use R commands to construct aligned event timelines for environmental, colonisation, and infection isolates.

Evaluate whether observed time and location overlaps make proposed reservoir-to-patient links plausible.

Use igraph commands to build and interrogate isolate networks containing genomic, spatial, and temporal edges.

Identify sampling bias, patient-movement gaps, and unsampled reservoirs that could confound transmission narratives.

Rank cross-niche candidates for hybrid-genome investigation using transparent genomic and epidemiological criteria.

Explain plasmid modular organisation and the limitations of replicon-based plasmid classification.

Run MOB-recon and inspect its evidence to separate candidate plasmid sequences from chromosomal sequence.

Distinguish autonomous plasmids from chromosomal integrations and unresolved repeat-linked contigs.

Run MOB-typer and PlasmidFinder to characterise plasmid replicons, relaxases, mating systems, oriT signals, and predicted mobility.

Run ISEScan and IntegronFinder and use curated similarity searches to annotate insertion sequences, transposons, and integrons around carbapenemase genes.

Diagnose plasmid reconstruction ambiguity caused by repeats, cointegrates, or multiple related plasmids.

Build a curated collection that separates complete plasmids, high-confidence partial plasmids, and unresolved sequences.

Run Mash to estimate pairwise plasmid similarity and explain how sketch size and distance thresholds affect clustering.

Use MMseqs2 commands to construct a plasmid protein-family presence-and-absence matrix.

Run clinker to produce and interpret linear gene-synteny comparisons of carbapenemase plasmids.

Run pyGenomeViz or BRIG to produce an interpretable circular comparison of complete plasmids.

Use minimap2 or nucmer commands to detect structural differences among related plasmids.

Use BWA-MEM2 and samtools commands to map isolate short reads to plasmid references and quantify breadth, depth, and ambiguous support.

Distinguish evidence for the same plasmid, a related plasmid backbone, and a shared carbapenemase cassette.

Test whether host-genome and plasmid relationships support clonal spread, plasmid spread, or both.

Identify near-identical plasmids occurring across Enterobacter and Citrobacter lineages, species, or sequence types.

Compare carbapenemase cassette conservation when plasmid backbones differ.

Assess the temporal and spatial opportunity for each proposed cross-niche plasmid-transfer event.

Evaluate whether an event is more consistent with environment-to-patient transfer, patient-to-environment transfer, or an unsampled common source.

Explain why genomic similarity alone rarely proves transfer direction or direct transmission.

Assign transparent confidence grades to proposed horizontal-transfer events.

Use R commands to integrate host phylogeny, plasmid clusters, species, niche, time, and location into an HGT evidence table.

Define evidence-based gene categories related to adhesion, biofilm formation, metal tolerance, disinfectant tolerance, and stress response.

Run BLAST or DIAMOND against a curated biofilm and persistence-gene reference set and filter matches using documented criteria.

Distinguish core lineage-associated biofilm genes from accessory or mobile genes.

Determine whether selected persistence genes are chromosomal, plasmid-borne, or located on other mobile elements.

Run pyseer to test niche-associated gene enrichment while controlling for bacterial population structure.

Interpret genomic detection of a biofilm-associated gene as functional potential rather than proof of phenotype.

Integrate biofilm and stress-tolerance findings into cautious hypotheses of damp-environment persistence.

Create an isolate-level audit table connecting read QC, assembly QC, taxonomic assignment, typing, and plasmid confidence.

Perform sensitivity analyses for genomic-distance, plasmid-similarity, and mapping thresholds.

Triangulate host-genome, plasmid, spatial, temporal, and niche evidence in selected case studies.

Design a coherent figure set combining cohort summaries, phylogenies, timelines, networks, and plasmid comparisons.

Write reproducible methods for the short-read, hybrid-assembly, population-genomic, and plasmid analyses.

Report results using language calibrated to the strength and directionality of the evidence.

Explain the effects of incomplete sampling, assembly uncertainty, and metadata limitations on the study conclusions.

Implement a rerunnable subset of the analysis with Snakemake or Nextflow and execute it from a sample sheet.