Hard to install igm

[liminghao663]

I tried to use igm in a ubuntu 20.04 system, but failure to install the dependent alabtools. I have to install the specific version packages instead of the latest to complied with alabtools.
Could you provide a detailed list of the version of the dependencies?

[liminghao663]

Another question, how to prepare the .hcs input file? Could this be converted from usually hic formats, such as .hic/.cool/.h5/ ?

[bonimba87]

In the past
conda install pandas swig cython cgal==4.14 hdf5 h5py numpy scipy matplotlib tornado ipyparallel cloudpickle
has worked fine, as long as cgal 4.14 is installed.
Can you please try to install each package separately and then share the error message you get when installing alabtools?

[bonimba87]

As for the .hcs input file... you definitely need a working alabtools package/environment to generate that. You can easily go from mcool to hcs input file as follows.

import alabtools, numpy, scipy
m = alabtools.Contactmatrix(ZZZ, resolution = XXX, genome = YYY)

where ZZZ = name of the .mcool file (string) , XXX = model resolution in kb (integer), YYY = genome segmentation (string), currently alabtools allows for 'hg19', 'hg38' (human) and 'mm9' (mouse) genome types.

m is an alabtools matrix object. m.matrix is a sparse (SSS) matrix, which can be converted to regular NumPy array by the command m.matrix.toarray().

Now, you can perform any normalization on A = m.matrix.toarray() you see fit (filtering, matrix balancing, etc) to prepare the input. A key issue is that you need to go somehow from contact frequencies (as in the mcool file) to contact probabilities, since the .hcs file only handles contact probabilities. After you are satisfied with your preprocessing, it is easy to store the updated/preprocessed (contact probability) NumPy array A by:

spm = scipy.sparse.csr_matrix(A)
m.matrix = alab.matrix.sss_matrix((spm.data, spm.indices, spm.indptr))

m.save('output.hcs')

The 'output.hcs' file has now the correct format to be fed to the IGM modeling pipeline. Hope this helps, I just added the 'from mcool to hcs HiC input' section to the repository to make it easier for people, thank you for bringing this up.

On a side note. We do have a preprocessing pipeline that is detailed in the SI to the Nature Methods paper, I am in the process of polishing and curating the code for carrying that out and will soon add that to the IGM repository as well, so stay tuned!

However, IGM works with any 'hcs' file that stores a contact probability matrix, and so can accommodate different preprocessing strategies according to necessity.

[ziyi10]

As for the .hcs input file... you definitely need a working alabtools package/environment to generate that. You can easily go from mcool to hcs input file as follows.

import alabtools, numpy, scipy m = alabtools.Contactmatrix(ZZZ, resolution = XXX, genome = YYY)

where ZZZ = name of the .mcool file (string) , XXX = model resolution in kb (integer), YYY = genome segmentation (string), currently alabtools allows for 'hg19', 'hg38' (human) and 'mm9' (mouse) genome types.

m is an alabtools matrix object. m.matrix is a sparse (SSS) matrix, which can be converted to regular NumPy array by the command m.matrix.toarray().

Now, you can perform any normalization on A = m.matrix.toarray() you see fit (filtering, matrix balancing, etc) to prepare the input. A key issue is that you need to go somehow from contact frequencies (as in the mcool file) to contact probabilities, since the .hcs file only handles contact probabilities. After you are satisfied with your preprocessing, it is easy to store the updated/preprocessed (contact probability) NumPy array A by:

spm = scipy.sparse.csr_matrix(A) m.matrix = alab.matrix.sss_matrix((spm.data, spm.indices, spm.indptr))

m.save('output.hcs')

The 'output.hcs' file has now the correct format to be fed to the IGM modeling pipeline. Hope this helps, I just added the 'from mcool to hcs HiC input' section to the repository to make it easier for people, thank you for bringing this up.

On a side note. We do have a preprocessing pipeline that is detailed in the SI to the Nature Methods paper, I am in the process of polishing and curating the code for carrying that out and will soon add that to the IGM repository as well, so stay tuned!

However, IGM works with any 'hcs' file that stores a contact probability matrix, and so can accommodate different preprocessing strategies according to necessity.

Hellow, I want to know whether the preprocessing code of this piece has been sorted out, which may be very necessary for the correct use of igm software.