lxsteiner / scg2024_day2_recruit

metagenomic read recruitment discussion and tutorial

Repo for metagenomic read recruitment lesson for BSCG course

Read recruitment or mapping is one of the most commonly used process in genome analysis. It can be used for various purposes but in single cell genomics , it is often applied to access the relative abundance of SAGs in metagenomic datasets (when metagenomes are used) or to estimate the expression level of the genes within individual SAGs (when metatranscriptomes are used).

In a nutshell short reads are aligned to a genomic reference sequence (many thanks to Meren for creating this animation)

This task is not as trivial as it sounds. Meren has an insightfull on the issue. Julia and I have also worked on the benchmarking individual mapping tools (use to be in the Supplementary material in Pachiadaki et. al., 2017 but now disappeared). It is now uploaded here.

CoverM

For this tutorial we will use CoverM, a pipeline that "aims to be a configurable, easy to use and fast DNA read coverage and relative abundance calculator focused on metagenomics applications". CoverM can calculate coverage of individual contigs or of genomes with coverm genome (the mode we will be using; detailed manual here). Calculating coverage by read mapping, its input can either be BAM (Binary Alignemnt Mapping) files sorted by reference, or raw reads and reference genomes in various formats.

CoverM offers the possibility to use two different aligners, bwa-mem or minimap2. The two aligners have comparible accurancies (with bwa-mem performing a bit better) but minimap2 runs 3-4 times as fast (Li, 2018). We will be using minimap2, which is the default aligner in CoverM.

Let's get started

• First let's clone this repository into your own working directory. To do so, open up a terminal window and navigate to your user lab space. Then clone the repo:

$ cd /mnt/storage/userlab/{your_username_here}/
$ git clone https://github.com/Bigelow-eSCG-tutorials/day2_recruitment.git
$ cd day2_recruitment

Use the folder panel the panel on the left to access your local notebooks. Let's start by initiating the mapping. While this (step 2) is running, I will demonstrate how I downloaded the metagenomic samples we are using (please do not try to run this here; you can use it as guide to download metagenomes relevant to your research).

• 1_Metagenome_Recruitment_Setup.ipynb: Commans we used to download the metagenomic reads we will use for the exercise. Julia's work (see supplementary material of Pachiadaki et al. 2017) showed that - regarding the estimation of relative abundance of genomes - mapping million reads provides comparible results with mapping larger metagenomes. For this reason, we are going to subsamples the genomes as we download them.
• 2_CoverM.ipynb: Run mapping using CoverM.
• 3_Metagenome_recruitment_plotting.ipynb: Explore and visualize the results.