genome: --genomeType SuperTranscriptome

[BenjaminDEMAILLE]

--genomeType SuperTranscriptome: condense the genome to the union of its annotated exons.

What changed

Overlapping exons are merged, the merged intervals concatenated, and the transcripts that overlap grouped into superTranscripts — one per condensed chromosome (st0, st1, ...). The index then covers only exonic sequence, and introns cannot be crossed because they are not there. Requires --sjdbGTFfile; --genomeType Transcriptome is still refused.

Four files are written alongside the index, as STAR writes them: superTranscriptSequences.fasta, transcriptSequences.fasta, superTranscriptSJcollapsed.tsv, and fullGenome/conversionToFullGenome.tsv.

The design decision that kept it small

condense returns the annotation already rewritten into condensed coordinates — seqname is the superTranscript, positions are 1-based within it, strand is +. The transcriptome tables, the junction extractor and the suffix-array builder then consume it unchanged; nothing downstream knows the genome was condensed. Genome::from_chromosomes is split out of from_fasta for the same reason: these chromosomes are built rather than read.

Divergence

DIVERGENCE.md §1.2, needing sign-off. STAR condenses the sequence before it fills the reverse-complement half of its genome buffer, so every minus-strand exon reads the allocation's spacer bytes and minus-strand superTranscripts come out as uninitialised memory. Here that half is already filled, so a minus-strand exon yields its actual reverse complement.

This one is not a corner case: it affects any annotation with minus-strand transcripts. A superTranscript built here holds real sequence where STAR's holds spacer bytes, so for those transcripts the two indices differ and alignments against them are not comparable. Reproducing STAR would mean writing code whose correct behaviour is to emit garbage, and a test asserting it.

One behaviour that looks like a gap and is not

A transcript's own intron leaves no junction behind. With nothing else annotated between two exons they end up contiguous in the condensed genome, so there is no gap left to record. A junction survives only when another transcript's exon separates them. Both cases have a test.

Verification

End to end (test_genome_type_supertranscriptome_condenses_to_exons): the 20 kb test genome with its two annotated 50 bp exons condenses to a single 100 bp st0; a read straddling the 200 bp intron aligns 50M at position 26; conversionToFullGenome.tsv names the two source intervals at 10000 and 10250.

Unit tests in src/genome/supertranscript.rs:

• exons_are_concatenated_and_the_intron_disappears
• the_annotation_comes_back_in_condensed_coordinates
• overlapping_exons_are_merged_once
• a_junction_is_recorded_when_another_exon_separates_the_two
• disjoint_transcripts_become_separate_supertranscripts
• minus_strand_supertranscript_is_reverse_complement_not_spacer — the divergence
• the_conversion_map_has_one_block_per_merged_interval
• a_gtf_with_no_usable_exon_is_an_error

Gate: 569 lib + 22 integration tests, cargo clippy --all-targets -- -D warnings, cargo fmt --check, MSRV 1.89 — all green.

Output-neutral for --genomeType Full, which is the default and every existing run.

Output neutrality, measured

10k yeast SE reads (CONTRIBUTING's dataset) through this branch and through origin/main, both at default flags: the SAM records are byte-identical (9724 each; the two files differ only in the @PG CL: line, which records the binary path).

Split out of #109 following the one-theme rule in CONTRIBUTING.md; #160 has the transform back-output and #109 keeps the libsais builder.