search

TaoYang-dev / hicrep

R package to evaluate the reproducibility of Hi-C data
25 stars 4 forks source link

About input Formal #60

Open werhoog opened 6 years ago

werhoog commented 6 years ago

Hi, thanks for this R packages. But I am confused about those input files. For example , I got contact matrix like: bin1 bin2 score * . ** . ****

as well as bed files to descript: chrom STAR END bin_number * . . . ** Both of them came from HiC-Pro software output. Therefore, how can I use those results to get such input files to Hicrep with intra-chromosome interaction?

Thanks!

TaoYang-dev commented 6 years ago

The new version uses n x n contract maps as input. The old version requires n x (3+n) as input, where the three additional columns are nothing but "chrom bin_start bin_end".

Please check the get.scc() documentation to figure out which version you are using.

-Tao

On Tue, Apr 10, 2018 at 10:48 PM, werhoog notifications@github.com wrote:

Hi, thanks for this R packages. But I am confused about those input files. For example , I got contact matrix like: bin1 bin2 score * . ** . ****

as well as bed files to descript: chrom STAR END bin_number * . . . ** Both of them came from HiC-Pro software output. Therefore, how can I use those results to get such input files to Hicrep with intra-chromosome interaction?

Thanks!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/MonkeyLB/hicrep/issues/60, or mute the thread https://github.com/notifications/unsubscribe-auth/ATEu6x7poKyeLaQ1oryd0BwYH0oPieMcks5tnW74gaJpZM4TPShO .

-- Tao Yang PhD Candidate Bioinformatics and Genomics Penn State University

werhoog commented 6 years ago

I got it.And as far as i am concernd, HiCrep is used by inner-chromosome interaction only. Would it be wrong if I merge all inner-chromosome interaction to a matirx.That is to say , suppose chromosome 1 may be a NN matrix ,and chromose 2 may be a (N-1)(N-1) matrix ?

Thank you!

TaoYang-dev commented 6 years ago

I am not sure what you mean by saying merging them together. I think you should perform the hicrep by chromosome, instead of combining them first and compute at one time.

-Tao

On Sun, Apr 15, 2018 at 8:51 AM, werhoog notifications@github.com wrote:

I got it.And as far as i am concernd, HiCrep is used by inner-chromosome interaction only. Would it be wrong if I merge all inner-chromosome interaction to a matirx.That is to say , suppose chromosome 1 may be a NN matrix ,and chromose 2 may be a (N-1)(N-1) matrix ?

Thank you!

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/MonkeyLB/hicrep/issues/60#issuecomment-381404253, or mute the thread https://github.com/notifications/unsubscribe-auth/ATEu6-ul_9uA024Zrwdib6rUtCnS4CbWks5to0JlgaJpZM4TPShO .

clagiamba commented 4 years ago

Hi, I am having the same problem starting with HiC-Pro input data: matrix=hicpro/hic_results/matrix/rep1/raw/500000/rep1.matrix bed=hicpro/hic_results/matrix/rep1/raw/500000/rep1_abs.bed

I cannot find the exact steps that will work. Are the steps to (1) first convert the matrix to full, e.g. using $HICPRO/utils/sparseToDense.py $matrix -b $bed -d ; (2) cbind the bed file, e.g. mat = cbind(bed[,c(1:3)], mat) ?

I am getting the error duplicate 'row.names' are not allowed... It would be really helpful if you could provide an example in the Vignette or reply here for everybody since HiC-Pro is so widely used. Many thanks