Public paired end chip seq dataset, both ChIP and input?

Question

[endrebak]

I want to add paired end support to my chip-seq software*, but to do that I need some paired end data. Do you know of any public PE chip-seq data? If you have some you could share a tiny subset of, that would also be greatly appreciated.

If you know a way to create fake PE bam files I'll accept that too.

• https://github.com/endrebak/epic

[GenoMax]

Here you go: http://www.ebi.ac.uk/ena/data/view/SRX1485421
http://www.ebi.ac.uk/ena/data/view/SRP062397
Human: http://www.ebi.ac.uk/ena/data/view/SRR1573750

[endrebak]

I have tried a few datasets from sra, but they were not PE, despite the labels. I'll look into those, thanks.

The last one looks interesting, but I could not find the input data (using mobile, that might be it).

[GenoMax]

No guarantee they are something you could use. They are labelled ChIP-seq and are PE :-)

[endrebak]

Yeah, those archive data are often terribly badly labeled. Takes ages to download and then you see that they cannot be used/aren't what you expected after all. Will perhaps try making my own, unless anyone has any datasets with both chip and input (only one of those has to be pe).

Score! The last file was input. Thanks! Ps. other answers still appreciated :)

Damn, with hisat2 you can just give the SRA-nb on the command line and it downloads and aligns it for you. Sweet: hisat2 -x hg38 --sra-acc SRR1573750 -S chip_SRR1573750.sam

Answer 1

[Ar]

You may try the paired end and single end data used in this paper. They did some comparison between paired and single end ChIP-Seq dataset.

[endrebak]

Could not find sequence read archive info in paper :/

[Ar]

Did you read the paper ? or saw where the dataset was uploaded ? The dataset are uploaded on GEO with GSE27679 id. Read the description for each sample. It says : Average pair-end library size: ~300bp . Go thru the paper and it is written quite explicitly about the paired end samples.

[endrebak]

Thanks, must have used the wrong cntrl-f queries :)

[Ar]

Good Luck with your project!

Answer 2

[jotan]

There's some good quality PE mouse ChIP seq in this H3K9me3 study

Answer 3

[Michele Busby]

Hi,

We used paired end for a bunch of histone modifications here:

http://biorxiv.org/content/early/2016/05/19/054387

I have not uploaded it to SRA yet.

You can email me if you want to use it mbusby at broadinstitute dot org and motivate me to do it if it is what you need. I think most of them are 2x36 but they are all paired end.

Paired end is better!

Michele

[endrebak]

I have what I need but thanks for offering! Please post an SRA link when you upload them.