Matt:LabNotes/2013-9-27

From ZhangLabWiki
Jump to navigation Jump to search

V4S1 MimicError Novoalign to Full Reference[edit]

  • Novoindex
 novoindex V4S1_full.ndx V4S1_full.fa
 # novoindex (2.8) - Universal k-mer index constructor.
 # (C) 2008 - 2011 NovoCraft Technologies Sdn Bhd
 # novoindex V4S1_full.ndx V4S1_full.fa 
 # Creating 16 indexing threads.
 # Building with 7-mer and step of 1 bp.
 # novoindex construction dT = 0.2s
 # Index memory size   0.001Gbyte.
 # Done.
  • Novoalign
 novoalign -d V4S1_full.ndx -f V4S1_mockseq_error_mimic.fq -F STDFQ -r ALL -o SAM -o FULLNW > V4S1_50onFull_novoalign.sam &
 #     Read Sequences:  2959000
 #            Aligned:  2953321
 #   Unique Alignment:  2949287
 #   Gapped Alignment:   395910
 #     Quality Filter:     4870
 # Homopolymer Filter:        0
 #       Elapsed Time: 461.655 (sec.)
 #           CPU Time: 7.4 (min.)
 # Done at Fri Sep 27 11:42:31 2013
  • Samtools
 samtools view -bS V4S1_50onFull_novoalign.sam | samtools sort - V4S1_50onFull_novoalign_sorted
 samtools view -h -F 4 -q 70 V4S1_50onFull_novoalign_sorted.bam > V4S1_50onFull_novoalign_sorted_filtered.sam
 samtools calmd -eS V4S1_50onFull_novoalign_sorted_filtered.sam /home/mzcai/DansProbes/V4S1_full.fa > V4S1_50onFull_novoalign_sf=.sam
 192 Errors (eg [bam_fillmd1] different MD for read 'Mock': '20^C0^C30' -> '20^CC30')
  • CountErrorNovoalign_perbase.pl does not work for this sam file because the read aligns somewhere from 28-33. I think since it is not revcomp that it's counting from the left end and since the probes aren't all uniform length, there's no way to do it this way
    • I'll align to V4S1_fullrevcomp that way I know it should start after the read primer region (19bp)