Skip to content

A Machine Learning Framework to Determine Geolocations from Metagenomics Profiling

Notifications You must be signed in to change notification settings

xmuyulab/metageopred

Repository files navigation

A Machine Learning Framework to Determine Geolocations from Metagenomics Profiling

Background

Studies on metagenomic data of environmental microbial samples found that microbial communities seem to be geolocation-specific, and the microbiome abundance profile can be a differentiating feature to identify samples' geolocations. We present a machine learning framework to determine the geolocations from metagenomics profiling of microbial samples.

Framework

The flowchart of the proposed framework

Data Downloaded and Preprocess

We downloaded the raw metagenomic sequencing data from the CAMDA 2019 Metagenomics Forensic Challenge provided by CAMDA in partnership with MetaSUB International Consortium. And the raw metagenomic sequencing data were preprocessed by fastp (v.0.19.4) for quality control including automatic filtering, trimming, and error removing.

About

A Machine Learning Framework to Determine Geolocations from Metagenomics Profiling

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •