hic (hi-chip) data loop calling with mango

[JungWoongJae]

Hi,
I'm a researcher in South Korea trying to use mango for Hi-ChIP data.

So I looked in to mango and guessed I should use stage 4~5 to call loops.

Is there a guide line I could use pre-aligned bam files (eg. HiC-pro output BAM files) and use mango stage4~5 to call loops?

It will be very helpful if I can get some advices or directions!.

Thank you very much.

Sincerely, WoongJae

[dphansti]

I have never used mango for Hi-ChIP data but I think it could be done. You need to first 'call peaks' where your TF or histone mark are bound/enriched. Then you could run just stage 5 of mango. You would need to provide two things:

1. A BEDPE file with all of your aligned paired end reads
2. A BED file describing the peak coordinates

For exact file formats see the "Intermediate Files" section of the README file here:
https://github.com/dphansti/mango

[JungWoongJae]

Thank you very much!!

I'll try and let you know

Thanks again.

[JungWoongJae]

I have never used mango for Hi-ChIP data but I think it could be done. You need to first 'call peaks' where your TF or histone mark are bound/enriched. Then you could run just stage 5 of mango. You would need to provide two things:

1. A BEDPE file with all of your aligned paired end reads
2. A BED file describing the peak coordinates

For exact file formats see the "Intermediate Files" section of the README file here:
https://github.com/dphansti/mango

Hi, dphansti

Sorry for the late response.

I tried to process mango as you directed me for hichip data.

To process mango, I went through steps below.

I made bedpe file using bowtie aligned bam file. (Which is aligned to restriction site cut reference genome)
I called peak with macs2 with the bam file without control file, and made slopPeak file with slopBed bedtools.
I made tagAlign file with the bam file using bamTobed -tag option.
And then I ran mango with the command,

>>Rscript mango.R --prefix $PREFIX --stages 5 --bedtoolsgenome ref.chrom.sizes

And the screen gave me below messages.

[1] "Starting mango ChIA PET analysis tool"
[1] "2019-01-31 14:29:32 KST"
[1] "bedtools version: bedtools v2.27.1"
[1] "macs2 version: macs2 2.1.2"
[1] "bowtie version: bowtie version 1.1.2"
[1] "counting reads per peak"
[1] "determining self-ligation distance"
Error in read.table(distancefile, header = FALSE, sep = "\t") :
  no lines available in input
Calls: calcDistBias -> read.table
Execution halted

It seems that I need .distance file.
How can I make .distance file?

Thank you

[bwubb]

@dphansti Greetings,

Any update to this issue? I am attempting to run the same analysis (Hi-ChIP input) and am experiencing the same error. It is not clear to me where the .distance file comes from or how I can generate one?

[dphansti]

You don't need to supply a distance file. It is probably not finding a different required file (though I am not sure why you wouldn't get an error for that.

Looking at the code more closely it looks like you need a correctly name .tagAlign, .rmdup.bedpe, and .slopPeak file for this to work. Do you have all of those?

Again I have not tested Mango on HiChIP data and I suspect that a more recently pipeline that was developed specifically for this might be more appropriate. You might want to consider FitHiChIP (https://www.biorxiv.org/content/10.1101/412833v1) which uses some of the same principles as Mango (Mango was derived in part from FitHiC) and was specifically designed for this type of data.

[bwubb]

@dphansti

Thank you for the reply and suggestions. My rmdup.bedpe is non-existent. Im fairly certain my input does not have duplicates, but by what criteria should I use, readnames? position?

My tagAlign file may be incorrect as well.

bamToBed -tag NM -i bwt2pairs.bam > mango.tagAlign