NMPFamsDB

NMPFamsDB catalogs protein families derived from metagenomic and metatranscriptomic data to support comparative, functional, and structural analyses across diverse microbial communities.

Key Features:

• Comprehensive Data Collection: Contains over 100,000 protein families, each comprising at least 100 members identified from metagenomic and metatranscriptomic analyses.
• Advanced Analytical Tools: Provides multiple sequence alignments and Hidden Markov Models (HMMs) for each family to enable comparative studies and functional prediction.
• Rich Metadata Integration: Annotates families with taxonomic information, ecosystem context, geolocation data, and predicted sequences and structures.
• Predictive Modeling: Includes 3D structure models predicted with AlphaFold2 for protein families to inform functional and interaction inference.

Scientific Applications:

• Repertoire Expansion: Expands known protein sequence clusters beyond reference genomes, doubling the number of recognized protein families.
• Proteome-wide Analysis: Enables proteome-wide comparative analyses across metagenomes and metatranscriptomes.
• Novel Function Discovery: Supports discovery of novel protein functions through sequence clustering, HMMs, and structural predictions.
• Ecological and Taxonomic Origin Exploration: Facilitates analysis of protein origins and taxonomy within native habitats using integrated taxonomic, ecosystem, and geolocation metadata.

Methodology:

Protein families are identified via metagenomic and metatranscriptomic analyses and bioinformatics pipelines that detect families lacking homologs in existing databases; families are represented by multiple sequence alignments and Hidden Markov Models, and 3D structures are predicted using AlphaFold2, with sequence analysis techniques applied for classification and functional annotation.