Scientific Methods — Pairwise Genome Comparison

Input & assembly statistics

Exactly two bacterial assemblies are compared. Statistics include total bases, contig count, N50, ambiguous bases and GC content over unambiguous bases.

Uncompressed .fasta, .fa or .fna bacterial assemblies. Each file is limited to 25,000,000 bytes, 100,000–20,000,000 sequence bases and at most 10,000 contigs. Validation checks non-empty records, unique record identifiers and IUPAC DNA symbols.

Average nucleotide identity

FastANI runs with assembly A as query and B as reference (-q / -r, 2 threads), using its algorithm defaults. The report records ANI, mapped fragments, total query fragments and mapped/total × 100 as aligned fraction. This fraction is query-fragment based, not whole-genome identity.

A missing reportable FastANI match is not zero ANI. ANI and its alignment support should be read together; AstroNexus does not turn this result into a clinical or automatic species-identification claim.

Annotation & protein comparison

Bakta annotates both assemblies (2 threads, --skip-plot, configured database). BLASTP compares A’s predicted proteins against B’s. Accepted alignments have identity ≥80%, query coverage ≥80%, subject coverage ≥80% and E-value ≤1e-5. Coverage is aligned coordinate span divided by the respective protein length.

Proteins with any qualifying alignment count as matched; the displayed best match per A protein is ranked by bitscore then identity. These are not guaranteed one-to-one orthologues. Pair-specific candidates have no qualifying match in the other submitted assembly; this does not prove unique function or biological absence.

Delivered outputs

The report/ZIP include ANI metrics, assembly statistics, annotation summaries, protein matches, pair-specific candidate tables, methods, tool versions, manifest and SHA-256 checksums. Source assemblies are excluded.