Closed lcolladotor closed 4 years ago
Code is running well memory-wise, but it's still being slow (a little over 1% done in 5 hrs). I can modify the window size from +- 1kb to say +- 250 bp or +- 500 bp if that helps (see https://github.com/LieberInstitute/brain-epigenomics/blob/master/meth_vs_expr/meth_vs_expr_QTL_near_nonCpG_meQTLs.R#L137). Though it would be a different window size than what we used for the rest of the features. With the current window size it's looking at 3,854,883 and 3,864,837 left and right side ends of the junctions respectively.
About 5% done after almost 24 hrs with the +1 1kb window.
Single end jobs are now done for those near nonCpG meQTLs.
4 runs total: