Running error about fithic HOT 7 CLOSED

Hi, seems like there may be a problem with the format of your input files. The dictionary structure supposed to have a list of fragments in it per each chromosome seems to be empty. This may be a problem with your fragment mappability file

from fithic.

Hi, thanks for the reply. The fragment file seems to be ok, as following:
########################
1 0 20000 621 1
1 40000 60000 1073 1
1 80000 100000 837 1
1 120000 140000 976 1
1 160000 180000 1793 1
1 200000 220000 846 1
########################
The inputs were produced by HiCPro2FitHiC.py.

Best,
Jing

from fithic.

what I can guess is that you have a chromosome (maybe chr Y if human) in your input file that has no fragments above mappability threshold so hence the error. Try removing chrs that are unmappable. If it doesn't work please share your input files. thanks

from fithic.

Hi,

I've checked that all chromosomes have the mappability threshold.
The inputs are too big (about 50Mb), so I put them on the google drive. https://drive.google.com/drive/folders/1_wm-CAsXY2oDzRGPLEuPtIIIT5UPIhqA. Please let me know if they are not available. Thanks a lot!

Jing

from fithic.

as I have suspected, there are scaffolds in your fragment list and bias file that do not have any valid/mappable fragment. See the examples below. if you don't remove them fithic won't work.
Scaffold_274 20000 -1
Scaffold_275 20000 -1
Scaffold_275 60000 -1
Scaffold_275 100000 -1
Scaffold_275 140000 -1
Scaffold_275 180000 -1

from fithic.

It's working! Thanks very much! I thought fithic would deal with those -1 value.
I have another question that how the ExpCC is calculated for inter-chr interaction, as I found some ExpCC values are equal to zero?

from fithic.

It is in the below piece of code. If bias values are outside the desired range then it gets set to zero.

        if allReg or interOnly:
            prior_p=interChrProb*(bias1*bias2)
            p_val=scsp.bdtrc(interxn.getCount()-1,observedInterAllSum,prior_p)
            interCount += 1
            # computing expected contact count
            if ((bias1 >= biasLowerBound) and (bias1 <= biasUpperBound) and (bias2 >= biasLowerBound) and (bias2 <= biasUpperBound)):
                expected_CC = (observedInterAllSum * prior_p)
            else:
                expected_CC = 0

from fithic.