---

name: file-io-and-format-conversion description: "Use Biopython SeqIO and AlignIO for offline sequence/alignment file parsing, writing, indexing, conversion, FASTA/FASTQ iterators, and compressed/BGZF tradeoffs." disable-model-invocation: true metadata: disco-role: operating license: NOASSERTION

File I/O and Format Conversion

Use this sub-skill when the task is to read, write, index, or convert biological sequence or multiple-alignment files with Biopython's Bio.SeqIO, Bio.AlignIO, low-level FASTA/FASTQ iterators, or BGZF-aware random access.

Route here for

• Choosing SeqIO.parse, SeqIO.read, SeqIO.write, SeqIO.convert, SeqIO.to_dict, SeqIO.index, or SeqIO.index_db.
• Choosing AlignIO.parse, AlignIO.read, AlignIO.write, or AlignIO.convert for multiple sequence alignment files.
• FASTA, FASTQ, QUAL, GenBank/GB, EMBL, Swiss-Prot/UniProt XML, tabular sequence files, PHYLIP, Clustal, Stockholm, NEXUS, MAF, and related file formats.
• Memory tradeoffs for streaming, materializing records, in-memory dictionaries, on-disk indexes, and raw record extraction.
• Low-level FASTA/FASTQ iteration when SeqRecord construction is unnecessary.
• Compression decisions: plain text handles, standard gzip for streaming, and BGZF for indexed random access.

Route elsewhere

• Sequence object semantics, annotations, features, coordinate behavior, translation, reverse complementing, and per-letter annotations: use sequence-objects-and-features.
• Search-result object models and BLAST/HMMER/PSL/SearchIO semantics: use alignment-search-and-phylogeny.
• Online downloads, Entrez/KEGG/ExPASy/UniProt service etiquette, and BioSQL databases: use web-databases-and-biosql.
• Structural file parsing for PDB/mmCIF objects: use structural-bioinformatics unless the task is only sequence extraction through a SeqIO format.

Operating references

• Start with references/file-io-workflows.md for task patterns, memory choices, indexing, conversion, and compression.
• Use references/format-reference.md for verified format names and read/write/index support.
• Use references/troubleshooting.md for common parser/writer/index errors and recovery steps.
• Run scripts/seqio_alignio_smoke.py for a safe offline sanity check of SeqIO, AlignIO, indexing, BGZF, conversion, and low-level iterators.

Guardrails

• Always pass explicit lowercase format names; Biopython does not infer the format from the filename extension.
• Prefer streaming (parse) for large files; only materialize into list or to_dict when the data size and memory budget are known to be small enough.
• Treat conversion as lossy unless the target format can store the same annotations, quality scores, alignments, and molecule type.
• Do not require network, external aligners, original examples, or repository test data for runtime recipes in this sub-skill.