how get TADs with a matrix from HiCPro?

[jacomegutierrez]

Hi, I want to analyse ice matrix from hicpro with the script matrix2insulation.pl, is this posible? because I got the ERROR: Must supply headered matrix.

[nservant]

Hi,
The HiC-Pro output cannot be directly loaded in matrix2insulation.pl
You have to convert it first.
I'm not 100% sure, but if I remember correctly matrix2insulation requires a dense matrix with bin names as rownames ... as in the my5C tool.
In the HiTC R package, you can load HiC-Pro data (using importC function) and export dense matrix using the export.my5C function.
I can also write a small tool to make the convertion in HiC-Pro if it helps.
Nicolas

[jacomegutierrez]

it would be great a small tool in HiCPro to convert matrix!

[jacomegutierrez]

I was trying to generated the input for matrix2insulation.pl in HiTC, first with importC then using export.my5C for all the matrix but the following error appear:

export.my5C(a, "15min_a_my5C", genome="BSgenome.Hsapiens.UCSC.hg19", per.chromosome=FALSE)
Start combining HTCexp objects ...
Object size: 77323496
Exporting data in '15min_a_my5C_hg19.mat' ...
Error in asMethod(object) :
Cholmod error 'problem too large' at file ../Core/cholmod_dense.c, line 105

Then I try the same per chromosome, but that matrix are not symmetrical!

[nservant]

Hi Jacome,
Will try to move on that asap.
Do you have somewhere an example of input file for matrix2insulation.pl ?
I have in mind a dense format with rownames/colnames as :
bin_number|organism|genomic_position| (ie. 35126|mm9|chr10:320001-360000)
Is that correct ?
Anyone knows if the organism is useful ? can I put a default to 'org' for instance ?
thanks

[jacomegutierrez]

Hi Nicolas, For matrix2insulation.pl there are two test files in https://github.com/dekkerlab/crane-nature-2015/tree/master/test/input The dense format you propose is correct, but I don´t know if the organism is necessary El vie., 1 mar. 2019 a las 9:03, Nicolas Servant (<notifications@github.com>) escribió:

…

Hi Jacome, Will try to move on that asap. Do you have somewhere an example of input file for matrix2insulation.pl ? I have in mind a dense format with rownames/colnames as : bin_number|organism|genomic_position| (ie. 35126|mm9|chr10:320001-360000) Is that correct ? Anyone knows if the organism is useful ? can I put a default to 'org' for instance ? thanks — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#9 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AbimGQIFTidciGTw19XX9wv39Yx7r3Jmks5vSUExgaJpZM4amZHZ> .

-- *Karina Jácome López*

[nservant]

Hi Jacome
I just released the sparse2dense.py script to convert HiC-Pro data to matrix2insulation.pl input.
See https://github.com/nservant/HiC-Pro/tree/devel/bin/utils
See --ins option
N

[jacomegutierrez]

Thank you so much! I generated the IS values without problems. El jue., 7 mar. 2019 a las 10:11, Nicolas Servant (<notifications@github.com>) escribió:

…

Hi Jacome I just released the sparse2dense.py script to convert HiC-Pro data to matrix2insulation.pl input. See https://github.com/nservant/HiC-Pro/tree/devel/bin/utils See --ins option N — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#9 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AbimGX36H8wFxSWGkW2vw4_aC6tCX0_Bks5vUToxgaJpZM4amZHZ> .

-- *Karina Jácome López*

[nservant]

great ! N

[hzaumsq]

When I use the hicpro result to look for Tad
1 .sparseToDense.py matrix/sample1/iced/40000/sample1_40000_iced.matrix -b matrix/sample1/raw/40000/sample1_40000_abs.bed -d -o 40000.matrix
the first step looks like correct;
2. perl /public/home/csxie/msq/tool/crane-nature-2015/scripts/matrix2insulation.pl -i 40000.matrix
ERROR: matrix must be symmetrical!
more 40000.matrix
Scaffold0 0 40000 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 291.48 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 86.158 0.0 0.0 0.0 0.0 0.0 113.291 215.461 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 350.491 0.0 0.0 0.0 0.0 352.66 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 81.84 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 104.334 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Then I got a error

[nservant]

Hi,
A couple of months ago, I added a '--ins" option to the spaseToDense script that should generate the right format for matrix2insulation.pl
Can you check if you used this option please ?
N

[guleiathms]

How much memory I should use for hg19, if I don't set -c? I used 120G, but it still runs out of memory.
Thanks.