Analyze Biological Data with Biopython

Biopython is a collection of Python tools for computational molecular biology. This skill organizes work across sequence handling, biological file conversion, alignments, database access, BLAST searches, structural analysis, phylogenetic trees, and additional analysis modules. It combines task selection with patterns that can be adapted to an actual dataset and environment. Bundled skill-relative reference guides cover the seven module areas; UPSTREAM-NOTICE.md records their pinned provenance, licenses, hashes, and import changes.

The workflow starts by identifying the biological question and relevant module, selecting the correct format and parameters, then combining modules where necessary. Typical sequences include fetching a GenBank record and analyzing it, searching with BLAST and retrieving selected hits, or constructing a phylogenetic tree from aligned sequences. These patterns require appropriate input data and validation; importing a module does not establish a scientifically valid result.

Use it for Python-based tasks involving DNA, RNA, proteins, sequence records, alignments, NCBI records, macromolecular structures, or evolutionary trees. Select methods according to the question, data quality, and assumptions rather than treating every available operation as appropriate for every dataset.

Check compatibility with the installed Biopython release. The overview's 1.85 version reference coexists with later compatibility notes, so it should not be read as a current-version guarantee. In particular, alignment defaults and deprecated interfaces can change reproduced results. External alignment tools, local BLAST, and DSSP require their own installations; the Python package does not supply every executable used in a workflow.

Inputs include sequence files or objects, accession identifiers, database queries, alignments, structure files, tree files, and analysis parameters. Outputs can be converted or filtered records, alignments, retrieved database records, parsed search hits, structural measurements, trees, annotations, statistics, and figures. Record format strings, scoring choices, thresholds, genetic codes, and database context so the analysis can be reproduced.

Bio.Seq represents sequences and supports complement, reverse complement, transcription, back-transcription, and translation, including alternative genetic-code tables and stopping at a stop codon. SeqRecord combines sequence data with identifiers, descriptions, annotations, features, and per-letter information. Bio.SeqIO provides iterator-based parsing, single-record reading, writing, and conversion across formats such as FASTA, FASTQ, GenBank, EMBL, SwissProt, and tabular records.

For large inputs, choose streaming parsing, an in-memory dictionary, a lazy index, or a persistent SQLite-backed index according to access and memory requirements. Low-level parsers can reduce overhead in high-throughput work. Compressed-file workflows need appropriate handles and format settings. Validate records before output and close file and index handles after use. FASTQ filtering uses quality scores; sequence extraction can preserve the record information needed downstream.

Bio.Align.PairwiseAligner supports global and local alignment with configurable scoring and substitution matrices, including BLOSUM and PAM families. The compatibility notes identify a default gap-score change in 1.86; set scores explicitly when reproducing older behavior. Bio.AlignIO handles multiple-alignment formats including Clustal, PHYLIP, Stockholm, aligned FASTA, NEXUS, EMBOSS, MSF, and MAF. Alignment tasks include constructing and extending alignments, sorting, gap handling, identity summaries, conserved regions, and protein alignment. Multiple-alignment operations require compatible aligned sequence lengths. Clustal Omega and MUSCLE are external command-line options.

Bio.Entrez accesses NCBI resources including PubMed, GenBank or Nucleotide, Protein, Gene, and Taxonomy. Configure an identifying email, obtain optional API credentials through the environment, respect provider limits, and handle network errors. Operations include database information, searches, summaries, record retrieval, linked records, uploaded identifier lists, global queries, and spelling suggestions. Select return types and text or XML modes for the database, parse responses accordingly, paginate large requests, and use search history where appropriate. Cache downloads and avoid repeated requests for unchanged records.

Bio.Blast supports NCBI web searches and local BLAST workflows. Choose nucleotide, protein, or translated search modes and a database that matches the question. NCBIXML parses XML results; tabular output offers another processing route. Filter hits using suitable E-values, identity, and other evidence, handle empty results, and retain search outputs. Batch searches, local databases, hit-sequence retrieval, ortholog investigations, and reciprocal-best-hit workflows require explicit search parameters. Large or repeated workloads may suit local execution or an alternative aligner rather than repeated remote queries.

Bio.PDB reads PDB and mmCIF structures through the Structure, Model, Chain, Residue, and Atom hierarchy. It supports coordinate access, distances, angles, dihedrals, neighbor searches, contact maps, transformations, superposition, RMSD, selected output, and sequence extraction. Structural workflows also address secondary structure and solvent accessibility through DSSP, Ramachandran information, missing atoms, binding-site neighborhoods, and center-of-mass calculations. Check alternative conformations, heteroatoms, multiple models, and structure quality. Suppressing parser warnings does not resolve the underlying data issue.

Bio.Phylo reads, writes, converts, navigates, analyzes, and draws trees in Newick, NEXUS, phyloXML, NeXML, and CDAO formats. Operations include clade searches, common ancestors, distances, statistics, subtree extraction, pruning, collapsing, ladderizing, rerooting, and topology comparisons. Tree-building patterns use distance matrices or alignments; consensus and bootstrap support help summarize collections of trees. Rich annotations and ASCII or matplotlib visualization serve different output needs. Validate taxon names, topology, branch lengths, rooting, and confidence values, and retain construction parameters and the input alignment.

Further modules address motif creation, consensus and position-weight matrices, motif searches and file exchange; GenePop population-genetics workflows, including Fst and Hardy-Weinberg calculations; sequence composition, molecular weight, melting temperature, GC skew, and protein properties; restriction sites and translation tables; clustering; and genome diagrams. SeqFeature supports annotated intervals and sequence extraction, while ambiguity codes and FASTQ qualities support sequence interpretation. Other patterns address PubChem records, open reading frames, codon counts, sequence complexity, and promoter extraction. Check strand, coordinates, pseudocounts, empty inputs, statistical assumptions, and the limits of each method.

Prefer PairwiseAligner over deprecated pairwise2 workflows. The compatibility notes identify removed Bio.HMM and Bio.MarkovModel interfaces and point HMM workflows toward hmmlearn; verify the relevant package APIs before adapting legacy code.

Install in a Python environment with NumPy available:

uv pip install biopython

Test on small datasets before scaling, cache intermediate results, investigate format and accession errors, and validate biological interpretations separately from successful script execution.

The skill is for Python users performing bioinformatics analyses who need to select modules, combine data-processing steps, and preserve the parameters and checks behind their results.