SOAPMetaS

SOAPMetaS profiles large metagenomic datasets using marker genes and Apache Spark for distributed computation.


Key Features:

  • Marker gene-based profiling: Utilizes marker genes to profile multiple samples simultaneously.
  • High performance and scalability: Processes 80 FASTQ samples totaling 416 GiB in approximately half an hour on distributed clusters.
  • Species-level accuracy: Achieves species profiling accuracy comparable to MetaPhlAn2.
  • Distributed computing compatibility: Runs on local setups, Spark standalone clusters, and YARN clusters and is executed via spark-submit on Apache Spark.

Scientific Applications:

  • Large-scale metagenomic profiling: Enables analysis of microbial community composition across large datasets.
  • "meph" mode: Recommended for analysis of novel microbial communities using an approach aligned with MetaPhlAn2.
  • "comg" mode: Suitable for samples from known microbial communities with a predefined gene set.

Methodology:

Implements marker gene-based profiling in Java and leverages Apache Spark for distributed execution, launched via spark-submit on local, Spark standalone, or YARN clusters.

Topics

Details

License:
GPL-3.0
Programming Languages:
Java, Python, Perl
Added:
1/18/2021
Last Updated:
2/20/2021

Operations

Publications

He S, Huang Z, Wang X, Fang L, Li S, Zhang Y, Zhang G. SOAPMetaS: profiling large metagenome datasets efficiently on distributed clusters. Bioinformatics. 2020;37(7):1021-1023. doi:10.1093/bioinformatics/btaa697. PMID:32766813.

PMID: 32766813
Funding: - Science Technology and Innovation Commission of Shenzhen Municipality: JCYJ20150831201123287