A multi-source domain annotation pipeline for quantitative metagenomic and metatranscriptomic functional profiling.

Area of Science:

• Microbiology
• Bioinformatics
• Genomics

Background:

• Traditional pathway modeling focused on single species, but microbiome sequencing reveals microbial communities' crucial role in biochemical functions.
• Understanding environmental impacts necessitates reconstructing biochemical pathways at the community level.
• Accurately quantifying gene-protein sequences and transcripts requires precise estimation of protein domain abundance in environmental samples.

Purpose of the Study:

• To develop a novel computational tool for improved functional annotation of microbial communities.
• To accurately identify protein domains and their abundance in metagenomic and metatranscriptomic data.
• To enhance the understanding of microbial interactions and environmental impacts.

Main Methods:

• MetaCLADE, a profile-based domain annotation pipeline utilizing a multi-source strategy.
• Direct application of the pipeline to sequencing reads for function identification.
• Application and comparison on simulated and diverse real-world metagenomic/metatranscriptomic datasets.

Main Results:

• MetaCLADE outperforms InterProScan in annotating protein domains across various environments.
• The pipeline demonstrates complementary predictions when combined with UProC, further improving functional annotation.
• Achieved high accuracy in annotating diverse environments including soil, marine ecosystems, ancient samples, and human tissues.

Conclusions:

• MetaCLADE significantly improves domain annotation for metagenomic and metatranscriptomic data.
• The tool facilitates the discovery of patterns in divergent sequences, crucial for understanding microbial community functions.
• Accurate functional annotation by MetaCLADE is vital for studying microbial interactions and environmental impacts.