This is a test version of Biostars. For the public version, visit https://www.biostars.org.
how to pool together biological replicates?

I have 2 biological replicates (Chip-Seq) and i want to know how to pool them together so that after pooling them together i may apply Homer to find peaks. I am doing it for the very 1st time thats y asking such a q.

Now my q is : I have found the peaks of replicates separately by Homer.

1.How to find the peaks higher than 1 read per million (rpm) from each replcate and then merge the peaks of replicates, So, how can I merge the peaks of replicates???????????????

2.Can I use Normalized tag count column to check for peaks higher than 1rpm.If yes,then what would be the interpretation of 145.7 normalized tag count. also what would be the value to check for >than 1rpm.

  1. how can i merge the peaks ?by mergeBed with -n option will be ok or not

Waiting for reply

chip-seq

I would also consider calling peaks on the separate replicates and then intersect the two peak sets to get a consensus.

3 answers

You can pool the reads and map them to get a single bam file. Alternatively you can merge bam files from the two replicates using samtools merge.

By the way, why would you want to merge the replicates? If the replicates are adequately sequenced, a better type of analysis would be to call peaks for the replicates separately and only keep the peaks that agree between the two replicates.

How can I keep the common peaks between replicates? how to do this ? I have no idea?

Or mergePeaks from HOMER.

you can either use cat function to pool all the fastq files or samtools merge to merge the bam files.

thanks alot.I have sam files. Can i use samtools merge function to merge the sam files?

I am not sure whether to use samtools merge as it help says only bam files. My safe bet would pool the fastq files and generate sam file with appropriate program

thanks alot.I have sam files. Can i use samtools merge function to merge the sam files?

Log in to answer this question.