FAME Metagenomics workshop 2022
Metagenomic binning is the process of separating sequences into bins representing different taxonomic groups to form draft genomes, generally known as metagenome assembled genomes. The most popular tools bin contigs as they are longer than reads and contain more information that can characterise the underlying organisms in the metagenomic sample.
In this session, we will use MetaBAT, a popular programs for binning metagenomes to bin our contigs.
We have already assembled the reads and provided the necessary assembly files to do binning. Let’s download these files and transfer them to the server.
Go to https://cloudstor.aarnet.edu.au/plus/s/2oFEKQv68UZoFS9.
Select all the files and click on Download.
The files will download to your laptop as assembly_binning.tar.
On your laptop, use WinSCP to transfer files.
If you are using the command line on your laptop, enter the following command and your password when prompted.
scp assembly_binning.tar username@115.146.84.253:~/.
You should see the progress of your upload.
Log in to the server using ssh command and enter your password when prompted.
ssh username@115.146.84.253
OR use putty or MobaXterm to login instead.
Once logged in, type the command ls. You should see assembly_binning.tar in the list.
assembly_binning.tarType in the following command to extract the files from assembly_binning.tar.
tar -xvf assembly_binning.tar && rm assembly_binning.tar
ls
If you type the command ls, you will see the folder assembly_binning.
Go into the folder assembly_binning and list the content using the following command.
cd assembly_binning
ls
You will see the following main files.
assembly_graph_with_scaffolds.gfacontigs.fastacontigs.pathsdepth.txtmetabat2_contig_bins.csvprep_result.pyNow we are set to bin our data using MetaBAT.
MetaBAT does the profiling based on tetranucleotide frequencies in the samples. It counts the 4-mers, (i.e. AAAA, AAAT, AAAG, AAAC, AATA, AATT, …) in the sequences and uses those to suggest which samples should go together. The advantage of MetaBAT is that it will return your contigs for you.
MetaBAT also uses coverage information of the contigs which is calculated from BAM files of every input sample. We need to map the individual reads to the contigs and get separate BAM files.
Since it takes a lot of time to generate the BAM files and create the depth file, we have already provided you with the depth file (depth.txt).
depth.txt file for your own dataIf you need to create BAM files for your own data, you can use the following command to create BAM files for each sample name listed in a file named file.txt.
mkdir bam_files
for FASTA in `cat file.txt`; do minimap2 -ax sr contigs.fasta $f_good_out_R1.fastq $f_good_out_R2.fastq | samtools sort > bam_files/$f.bam done
Now we can generate a depth file from the BAM files using the command jgi_summarize_bam_contig_depths provided from MetaBAT. This goes through the BAM files and calculates the average number of reads covering each base of the contigs. You can run the following command to generate the depth file.
jgi_summarize_bam_contig_depths --outputDepth bam_files/depth.txt *.bam
If you enter head depth.txt you will see the following output.
contigName contigLen totalAvgDepth SRR1237780.bam SRR1237780.bam-var SRR1237781.bam SRR1237781.bam-var SRR1237782.bam SRR1237782.bam-var SRR1237783.bam SRR1237783.bam-var
NODE_1_length_110371_cov_8.539967 110371 20.851 0.535061 3.59679 0.0114134 0.0112765 18.813 41.6837 1.49151 2.26044
NODE_2_length_60492_cov_10.103049 60492 24.3014 0.0307083 0.0435139 0.00884956 0.00876961 22.3116 22.6173 1.95033 2.2689
NODE_3_length_57035_cov_7.080186 57035 21.9248 2.48215 2.94401 0.804096 0.81977 15.9555 16.8353 2.68306 3.23609
NODE_4_length_54641_cov_7.481717 54641 23.0121 2.11929 2.316 0.465783 0.547674 16.9546 28.0057 3.47245 6.37836
NODE_5_length_54014_cov_16.257900 54014 44.0697 3.03223 3.37595 2.11247 2.12222 35.9221 44.55 3.00295 3.1778
NODE_6_length_48409_cov_10.290069 48409 24.9984 0.0504362 0.0953332 0.0121014 0.015298 22.8362 22.5818 2.09973 2.32797
NODE_7_length_47877_cov_7.628414 47877 23.3906 2.46823 2.47848 0.97712 1.0791 17.2922 22.9389 2.65305 3.35134
NODE_8_length_46824_cov_7.158524 46824 22.1051 2.49087 2.48935 0.862172 0.872394 16.2341 18.6561 2.51791 3.07479
NODE_9_length_46478_cov_5.442970 46478 20.1159 0.621849 0.710255 0.233876 0.256314 12.0661 13.6028 7.19403 7.6685
The rows represent contigs and columns represent further information about the contigs and coverage in each sample.
Let’s bin our contigs using MetaBAT. We will use MetaBAT2 which is the latest version of MetaBAT.
metabat2 -i contigs.fasta -a depth.txt -o metabat_bins/bin
You can find the bins identified by MetaBAT2 in the folder metabat_bins.
GraphBin is a metagenomic contig bin refinement tool that makes use of the contig connectivity information from the assembly graph to refine bins and recover contigs discarded.
Once we have our bins from MetaBAT2, we can use GraphBin to refine the MetaBAT2 binning result.
We have to format our binning result so that GraphBin can read the binning result. We have provided a script named prep_result.py which you can find in the bundled data. You can run it as follows.
python prep_result.py --binned metabat_bins --output ./
Since, we used metaSPAdes to assembly out data, we can run the metaSPAdes version of GraphBin as follows.
graphbin --assembler spades --graph assembly_graph_with_scaffolds.gfa --contigs contigs.fasta --paths contigs.paths --binned initial_contig_bins.csv --output ./