Biotite

1.6.0 · active · verified Thu Apr 09

Biotite is a comprehensive Python library (current version 1.6.0) for computational molecular biology, offering a broad set of tools for sequence analysis, structural bioinformatics, and accessing data from biological databases. It leverages NumPy arrays for efficient, high-performance operations and provides seamless interfaces to integrate with external bioinformatics software, allowing users to streamline their analyses from basic scripting to developing full software packages. The library maintains an active development and release schedule, with significant updates in recent years.

Warnings

breaking As of Biotite v1.6.0, the `biotraj` package is now a mandatory dependency for trajectory file interfaces in `biotite.structure.io`, and `mdtraj` is no longer required for this purpose. Projects relying on `mdtraj` through Biotite's internal interfaces might require adjustment.
Fix: Ensure `biotraj` is installed (`pip install biotraj`) and update import paths if directly using `mdtraj` features that were previously proxied by Biotite.
gotcha Biotite internally stores most sequence and structure data as NumPy `ndarray` objects. While offering high performance and intuitive NumPy-like indexing, users accustomed to other bioinformatics libraries (e.g., Biopython) might need to adapt to this NumPy-centric data model.
Fix: Familiarize yourself with NumPy array operations and indexing. Biotite's documentation provides examples of how to interact with its NumPy-based data structures.
gotcha Biotite is organized into several subpackages (e.g., `biotite.sequence`, `biotite.structure`, `biotite.database`). Specific functionalities reside within these submodules, requiring explicit imports from the relevant subpackage rather than a single top-level `import biotite`.
Fix: Always refer to the official documentation or example gallery to find the correct import paths for the specific classes or functions you intend to use.
deprecated In `biotite.sequence.graphics`, the default color scheme for visualizing sequence alignments changed from `rainbow` to `flower` in v1.6.0. The `flower` scheme is considered to represent amino acid similarity more effectively.
Fix: Explicitly set `color_scheme='rainbow'` if you wish to retain the old default, or adapt to the new `flower` default. Consider if your visualization interpretations are affected by the color scheme change.

Install

pip install biotite PyPI
conda install -c conda-forge biotite Conda

Imports

ProteinSequence

from biotite.sequence import ProteinSequence

entrez
```
from biotite.database import entrez
```

FastaFile

from biotite.sequence.io.fasta import FastaFile

align_optimal

from biotite.sequence.align import align_optimal

Quickstart

Downloads two protein sequences (avidin and streptavidin) from the NCBI Entrez database, parses them from a FASTA file, and performs a pairwise optimal sequence alignment using the BLOSUM62 matrix with affine gap penalties.

import biotite.sequence.align as align
import biotite.sequence.io.fasta as fasta
import biotite.database.entrez as entrez
import os

# Download FASTA file for the sequences of avidin and streptavidin
# The 'file_name' should ideally be a path to a temporary file.
# For a runnable example, we'll use a simple name and ensure cleanup in a real scenario.
file_name = "sequences.fasta"
uids = ["CAC34569", "ACL82594"] # Example UIDs for avidin and streptavidin
entrez.fetch_single_file(
    uids=uids,
    file_name=file_name,
    db_name="protein",
    ret_type="fasta"
)

# Parse the downloaded FASTA file and create 'ProteinSequence' objects
fasta_file = fasta.FastaFile.read(file_name)
avidin_seq, streptavidin_seq = fasta.get_sequences(fasta_file).values()

# Align sequences using the BLOSUM62 matrix with affine gap penalty
matrix = align.SubstitutionMatrix.std_protein_matrix()
alignments = align.align_optimal(
    avidin_seq, streptavidin_seq, matrix,
    gap_penalty=(-10, -1),
    terminal_penalty=False
)

print(f"Number of alignments: {len(alignments)}")
if alignments:
    print("First optimal alignment:")
    print(alignments[0])

# Clean up the downloaded file
os.remove(file_name)

view raw JSON →