LieberInstitute / BiocMAP

Our typical analysis steps for WGBS data, starting from FASTQ reads and ending with the construction of bsseq objects. Created using nextflow
http://research.libd.org/BiocMAP/
2 stars 0 forks source link

Fix smoothing of CpG bsseq objects #4

Closed Nick-Eagles closed 3 years ago

Nick-Eagles commented 4 years ago

The current approach is to write the final 2 merged bsseq objects to a single directory, using the HDF5Array package to give each object a separate prefix ("CpG" and "CpH") when storing assays in .h5 files. However, bsseq currently does not correctly smooth objects given a prefix, so the final "CpG" bsseq object is not smoothed as desired.