search

dekkerlab / cworld-dekker

perl cworld module and collection of utility/analysis scripts for C data (3C, 4C, 5C, Hi-C)
Apache License 2.0
63 stars 20 forks source link

no data available error #14

Closed Yijun-Tian closed 2 years ago

Yijun-Tian commented 2 years ago

Hi Bryan and Sergey, When I run changeMatrixHeaders.pl, error message shows below:

perl /software/cworld-dekker/scripts/perl/changeMatrixHeaders.pl -i COLUMN.score.matrix.gz --hmf Bin.MAP --verbose 1

Tool: changeMatrixHeaders.pl Version: 0.41.1 Summary: replace matrix row/col headers, or subset matrix by list of headers

building matrix object [] checking headers ... [valid] parsing headers ... [y=5723, x=5723, xy=11446] building headers objects ... [done] parsing contigs ... [y=1, x=1, xy=1] calculating header overlap ... [done] checking symmetry ... [no] identifying NA rowcols ... [0] calculating optimal sparsity ... [0]

inputMatrixName COLUMN.score inputMatrixPath /mnt/raid0/Yijun_Tian/ATAC_GEX/cicero-HiC/ matrixHeaderFlag 0 equalHeaderSizing 1 equalHeaderSpacing 1 headerSizing 0 headerSpacing 0 binningStep 0 equalHeaderFlag 1

contigs 1

yHeaders 5,723

xHeaders 5,723

totalHeaders 11,446

assembly NA numNArowcols 0 symmetrical 0 numInteractions 32,752,729 [midpoint distance mode] num_zeros 32,508,206 num_nan 0 num_nonzero_nonnan 244,523 missingValue 0

loading matrix data ... getData 0.00% complete (1/5723)... loaded 0 datapoints no data available! at /usr/local/share/perl5/cworld/dekker.pm line 1758. cworld::dekker::getData("COLUMN.score.matrix.gz", HASH(0x14215e8), 1) called at /software/cworld-dekker/scripts/perl/changeMatrixHeaders.pl line 186

It seems in the "building matrix object" phase, my matrix is judged as non-symmetrical. So I ran matrix2symmetrical as below:

perl /software/cworld-dekker/scripts/perl/matrix2symmetrical.pl -i COLUMN.score.matrix.gz -v

Tool: matrix2symmetrical.pl Version: 0.41.1 Summary: transform rectangular matrix into square (symmetrical) matrix

building matrix object [] checking headers ... [valid] parsing headers ... [y=5723, x=5723, xy=11446] building headers objects ... [done] parsing contigs ... [y=1, x=1, xy=1] calculating header overlap ... [done] checking symmetry ... [no] identifying NA rowcols ... [0] calculating optimal sparsity ... [0]

inputMatrixName COLUMN.score inputMatrixPath /mnt/raid0/Yijun_Tian/ATAC_GEX/cicero-HiC/ matrixHeaderFlag 0 equalHeaderSizing 1 equalHeaderSpacing 1 headerSizing 0 headerSpacing 0 binningStep 0 equalHeaderFlag 1

contigs 1

yHeaders 5,723

xHeaders 5,723

totalHeaders 11,446

assembly NA numNArowcols 0 symmetrical 0 numInteractions 32,752,729 [midpoint distance mode] num_zeros 32,508,206 num_nan 0 num_nonzero_nonnan 244,523 missingValue 0

loading matrix data ... getData 100.00% complete (5724/5723)... loaded 32,752,729 datapoints done

symmetriciz'n data... symmetrical error, matrix appears symmetrical (0 vs 0)! at /software/cworld-dekker/scripts/perl/matrix2symmetrical.pl line 152.

It said my matrix is actually symmetrical. I am pretty sure that my x and y headers are identical, but the headers parsing results are also strange. Lastly, the changeheaders error only happend for chromosome 13 to 22 and chromosome Y. Do you have any idea about the reason?

Thanks

Yijun-Tian commented 2 years ago

Hello it somehow returned to normal for chromosome 13 to 22 today.