nfcore proteinfamilies

nf-core/proteinfamilies generates nascent protein families or assigns new proteins to existing families from amino acid sequence data.

Key Features:

• Scalable architecture: Resource usage scales approximately linearly with input size.
• Open-source implementation: Built as a parametrizable Nextflow pipeline.
• Family generation: Constructs novel protein families from uncharacterized sequences.
• Family assignment: Assigns new sequences to existing protein families.
• Multiple sequence alignment integration: Supports inspection of conserved residues to provide functional clues.

Scientific Applications:

• Functional annotation transfer: Groups uncharacterized metagenomics-derived sequences with experimentally characterized proteins to enable annotation transfer.
• Protein function prediction: Uses conserved residues from multiple sequence alignments to infer function in the absence of existing functional information.
• Microbial diversity analysis: Processes metagenomics-derived amino acid sequence data to study microbial diversity and evolutionary relationships.
• Database benchmarking: Generates protein families that closely resemble manually curated families in widely used databases.

Methodology:

The pipeline is implemented in Nextflow and uses computational benchmarks to demonstrate linear scaling with input size, and biological benchmarks to validate that generated families resemble manually curated database families.