Kun:LabNotes/MONOD/2014-1-17

From ZhangLabWiki
Jump to navigation Jump to search

MONOD Round 2[edit]

Target identification[edit]

  • Data use:
    • Cancer data: same as Round 1
    • Placenta data: ENCODE RRBS (BC_Placenta_UHN00189); WGBS GSE39775 (I downloaded the raw reads and repeated the mapping to hg19)
    • Whole blood data: Round 1 data, plus 32 UCLA_SZ BSPP330k data.
 ./find_fetal_DMS_v2.pl > fetal_DMS_v2.txt
 ./find_DMS_PANC_GBM_v2.pl > PANC_GBM_DMS_v2.txt
 ./extract_clusters.pl fetal_DMS_v2.txt > fetal_DMS_clusters_v2.txt
 ./extract_clusters.pl PANC_GBM_DMS_v2.txt > PANC_GBM_DMS_clusters_v2.txt

Probe design[edit]

  • I designed probes for both the plus and minus strands to achieve a better coverage. In addition, I added a requirement that no CpG is allowed in H1/H2 annealing region to avoid methylation dependent capturing bias. Gap size: 125-175bp.
  • Probe filtering and selection.
    • Fetal probes: 173,034 probes were designed for 54,660 targets (39,517 targets failed). After filtering probes that have less than 4 CpG sites in the captured regions, I got 159,874 probes. I kept all probes in chr8,9,13,18,21,22; for the other chromosomes, randomly sample enough to fill the order.
    • GBM_PC probes: 23,707 probes were designed for 13,181 targets. After removing probes that either overlap with the fetal probe set or having fewer than 4 CpG sites in the captured region, I obtained 18,104 probes.
  • In total, I included 18,104 probes for GBM_PANC, 65343 for the fetal DMR in the order. I used the 8bp UMI. These oligos are in the DMR540k set F, and should be amplified by the V6 primer set.

Capture & sequencing[edit]

  • Noi performed the experiment, and the sequencing libraries were sequenced in the 140310_MiSeq run.

Data analysis[edit]

# reads # mapped reads % mapping # on-target reads % on-target # unique reads Clonal rate
1,346,769 937,785 69.60% 35724 3.8% 35521 0.6%
1,092,716 716,930 65.60% 28539 4.0% 28441 0.3%
1,172,030 786,617 67.10% 27690 3.5% 27641 0.2%
1,291,158 864,157 66.90% 34947 4.0% 34781 0.5%
1,110,997 687,345 61.90% 27167 4.0% 27045 0.4%
1,243,264 811,064 65.20% 32016 3.9% 31963 0.2%
1,425,464 880,798 61.80% 36412 4.1% 36209 0.6%
2,172,772 1,476,496 68.00% 64276 4.4% 63603 1.0%
1,345,928 757,309 56.30% 26866 3.5% 25971 3.3%
1,307,927 808,708 61.80% 33887 4.2% 33216 2.0%
1,906,290 1,070,550 56.20% 39475 3.7% 36852 6.6%