Ylaine/2009-7-13

From ZhangLabWiki
Jump to navigation Jump to search

False Negatives[edit]

  • Continue debugging code from Friday
    • Split SNP file into 4 1M-line files; 48759 in target region
    • Search for locations in 'noseven.chrlocs' using 'grep' within 'system'
my $command = "grep 'chr$chr\t$loc' $locsfile";
system($command);
  • MAQ/MAQ default SNP calling returned 8376 SNPs, 5906 of which have coverage >= 8x
  • Use 'grep' -f to search for locs in MAQ results:
grep -c -f covered.targetSNPsaa ../testout.txt
  • Found 6503 SNPs in 8x covered region. 4275 also found by MAQ (false negative rate of 34.26%)

BWA/SAM[edit]

  • Compare results using default filter and filter written by Dr. Zhang to results obtained with MAQ
  • Files are:
NA12878_40bp_061109.sequence.pileup.varFilter.txt
NA12878_40bp_061109.sequence.pileup.variants.txt
  • 9077/14996 of SNPs in 'variants' at dbSNP locations
    • 1K Genome comparison: (edited 7/15/09)
1K Genome In dbSNP Percent Not in dbSNP Percent
Match 7836 86.33% 222 3.75%
Miss 752 8.28% 5662 95.66%
Mismatch 489 5.39% 35 0.59%
Total 9077 5919
    • Using method from Friday, FP rate is 34.49%
    • Average quality comparison (not sure what the columns are):
' Col 5 Col 6 Col 7 Col 8
In dbSNP 74.25 81.56 36.04 12.41
s.d. 54.59 58.25 3.01 11.58
Not in dbSNP 39.30 40.64 34.50 10.79
s.d. 26.06 27.77 4.58 10.12
  • According to website, columns should be consensus quality, SNP quality, maximum mapping quality, and read number (coverage)
http://samtools.sourceforge.net/cns0.shtml