Hi, I am using Picard CollectHsMetrics tool to obtain a bunch of statistics about the regions in my targeted panel such as mean coverage and mean coverage for each interval. Since the panel is small, there is a huge number of duplicate reads and CollectHsMetrics only takes into account the unique reads and so the reported mean coverage is smaller than it actually is. Is there a switch that I can turn on/off to take into account all reads (incl duplicate reads) and not just unique reads.
Here is how I run it:
java -jar picard.jar CollectHsMetrics I=sample.bam O=sample.realigned.metrics TI=intervals.bed PER_TARGET_COVERAGE=sample.target_region.coverage BI=intervals.bed REFERENCE_SEQUENCE=hg19.fa
BTW, the input file (sample.bam) has gone through the step of duplicates marking. Even if I run CollectHsMetrics before marking duplicates, the results are the same.
Any help will be appreciated. Thanks!