Kun:LabNotes/MONOD/2015-7-6
Jump to navigation
Jump to search
Data analysis: re-defining bins based on methylation haplotype blocks
- I have previously defined a set of methylation LD blocks based on the N37 and whole-blood WGBS data, plus tumor RRBS data:Kun:LabNotes/MONOD/2015-3-11.
- Since several WGBS data sets have been published lately, I decided to repeat this using all WGBS data, in order to be as unbiased as possible.
- The list of WGBS data sets included are (Dinh did all the mapping and gave me the mapped bam files):
- Whole blood WGBS data (REF);
- Our internal N37 WGBS data (ten tissues): /media/LTS_33T/WGBS_LTS33/Hg19/Noi_N37_WGBS/BAMfiles
- WGBS data from H1 ESCs and differentiated cells (REF): /media/LTS_33T/WGBS_LTS33/Hg19/Xie2013/Re-map/BAMfiles
- WGBS data from human adult tissues (Salk, REF): /media/TmpStore2/Dinh_Store2/Tissues_WGBS_Ecker/BAMfiles
- WGBS data from a cancer cell line and a colon cancer tumor tissue (REF): /media/TmpStore2/Dinh_Store2/Tumor_WGBS/BAMfiles
- I still use the same "mappable bins" to generate the initial hapInfo files for mld_block partitioning. This was done on TSCC since there are so many files to process.
- Transfer bam files to TSCC.
- Create a "sampleInfo" file to each data set.
- Create job files and submit them to the queue.
/home/k4zhang/my_oasis_tscc/MONOD/batch_bam2hapInfo2.pl Media:H1ESC_WGBS_sampleInfo_mappable_bins.txt /home/k4zhang/my_oasis_tscc/MONOD/batch_bam2hapInfo2.pl Media:Ecker-Tissue_WGBS_sampleInfo_mappable_bins.txt /home/k4zhang/my_oasis_tscc/MONOD/batch_bam2hapInfo2.pl Media:N37_10_tissue_pool_WGBS_sampleInfo_mappable_bins.txt /home/k4zhang/my_oasis_tscc/MONOD/batch_bam2hapInfo2.pl Media:whole_blood_WGBS_sampleInfo_mappable_bins.txt /home/k4zhang/my_oasis_tscc/MONOD/batch_bam2hapInfo2.pl Media:tumor_WGBS_sampleInfo_mappable_bins.txt
- Transfer all hapInfo files to genome-miner, merge and call mld_blocks.
./merge_all_WGBS_hapInfo_by_chr.pl
- Transfer the bed files to TSCC, and generate all hapInfo files for the mld_blocks.