Question about multiple fastq files from the same library or from the same biological replicate

[Savio Chow]

Hi,

Thank you for the great response on this forum! I have three questions regarding multiple fastq files which describes different scenarios:

1) If different pairs of fastq files are from the same library, then I should place all these fastq pairs into the same fastq folder to run juicer.sh at the same time and it should combine all these fastq files from the same library into one result?

2) If one library only has one pair of fastq files, the folder should only contain this pair (if Q1 is correct). Although one biological replicate can be used to generate multiple libraries, we should not mix fastq files from multiple libraries of the same biological replicate. After running juicer.sh on each library, we can then concatenate the merged_nodups.txt from different libraries generated from the same biological replicate to merge all the libraries, is this correct?

3) If a cell line has multiple biological replicates, we can use mega.sh to merge all its biological replicates to form the final per cell line hic data, is this true?

So, if we have one cell line with biological replicate 1 and 2, each of these replicates has 3 libraries, my workflow would be: 1) for each library (6 in total), run juicer.sh (will either only have 1 pair of fastq files, or multiple pairs), 2) concat the output of 3 libraries for biological replicate 1 and 2 respectively, getting 2 merged files in total, 3) run mega.sh on these 2 merged files so that we have one file per cell line.

Best wishes,

Savio Chow

[Moshe Olshansky]

Hi Savio,

First of all, the answer to your questions 1-3 is yes (to all of them).

As to your final question. after you get 6 merged_nodups.txt files, you can make 6 separate hic files or you can merge each 3 files from the same replicate and produce two hic files (a file for each biological replicate) or you can merge all 6 and build one hic files for the cell line. These 3 options are no mutually exclusive. What you choose depends on what you are planning to do next.

Best regards,

Moshe.

[Wenjun Yan]

Hi Moshe,

I'm in a similar situation, and I'm trying to merge the 2 replicates. Could you describe more on how to move forward? I have .hic files from both samples, I also merged the merged_dedup.bam to have the pooled_merged_dedup.bam using samtools merge. I tried to run juicer.sh on the pooled_merged_dedup file with -S final, but it failed because merged1.txt/merged30.txt doesn't not exist (I only have merged_dedup.bam solft linked to pooled_merged_dedup.bam in the aligned folder). I'm using juicer_tools.2.20.00. Thank you!

Best,

Wenjun

[Moshe Olshansky]

Hi Wenjun,

I haven't touched Juicer for more than a year. 

One way is to have that pooled_merged_dedup.bam and all the required files and run the pipeline from that stage. Another possibility is to look at juicer.sh and run the commands manually starting from that pooled file. I am not sure which way is easier. Of course you could run mega.sh with the two experiments/replicates, but this will mean repeating all the time-consuming steps.

Sorry that I can not help more.

Good luck,

Moshe.

--
You received this message because you are subscribed to a topic in the Google Groups "3D Genomics" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/3d-genomics/Cu3YoG2gV50/unsubscribe.
To unsubscribe from this group and all its topics, send an email to 3d-genomics...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/3d-genomics/dad93027-5f09-4a5a-ac3e-1a540a3cb0cfn%40googlegroups.com.