Dear colleagues,
I pooled together dozens of samples into one library, and sequenced them together. This yielded millions of reads, with the sequencer assigning each unique read-IDs. While the samples have thousands of reads, each read ID corresponds to only one sample.
For example, the read-ID 1RDD4_00016_01732 (first column) only belongs to the sample o120 (second column).
1RDD4_00016_01732 o120
1RDD4_00016_01756 o297
1RDD4_00016_01943 o316
1RDD4_00016_02031 o296
As shown above, the read IDs are paired with their respective samples, and listed in a group-file format (i.e., two columns, tab-separated; https://www.mothur.org/wiki/Group_file).
With a group-file, and fasta file, how can I use Unix (or a similar program) to:
Best,
Gary Sur
1) You must have barcoded your pooled samples, thus the resulting fastq.gz files must be according to the barcodes.
2) You should use any specific pipeline which can handle the data according to the manual
3) If you have already created a mess with all fasta together, you could seperate the fasta sequences, but what will you do with that ? Than you might be asking to merge and make OTUs / OTU table ?
Although, there is some confusion, what kind of fasta file you have, but what i get from the asked question, you can easily make theme fasta files, by some tricks:
1RDD4_00016_01732 o120
1RDD4_00016_01756 o297
1RDD4_00016_01943 o316
1RDD4_00016_02031 o296
i assume these columns are tab separated.
1) open the file in vi text editor.
2) type %s/^/>/g
this will result like
>1RDD4_00016_01732 o120
>1RDD4_00016_01756 o297
>1RDD4_00016_01943 o316
>1RDD4_00016_02031 o296
3) now type
%s/\t/\r/g
this will result like
>1RDD4_00016_01732
o120
>1RDD4_00016_01756
o297
4) a simple script in perl or any other language can easily differentiate them into individual file by splitting by ''>"
Dear Abhijeet Singh and Salman Sadullah Usmani,
Thank you both for your help as well as inspiration for what I need to do. I am using MEGAN6 to visualize and analyze my data. I wanted to see if there was a unix pipeline to take the MEGAN6 output, and separate the many samples with their sequences into separate FASTA files. With separate fasta files, I was thinking I could compare and contrast the each sample's OTUs.
But I now realize that's impossible, and/or more trouble than it's worth. Instead, I've returned to my original fasta file, and used mothur's split.groups command to get what I need.
Best,
Gary
Gary Sur I appreciate your open mind to get a way out of the discussion. It is indeed possible to do what you asked in question, but what would be the purpose. If you just have to compare the samples which ofcourse every microbiobe/sequencing study does, use programs like phyloseq/vegan/ampvis etc. to use the sequencing results/OTU/OTU table and make something meaningful out of it.
- Badges
- Science method