How to best use my lab data to check my alignment??

[lg36]

Dear All,

These are three different pipelines I've used to call SNPs on the same samples:

1) Raw FastQ File --> Direct Mapping to Reference --> SNP discovered 1,200

2) Raw FastQ File --> De Novo Assembly --> Extract Paired Contigs --> Map Paired Contigs to Reference --> SNPs discovered 1,089

3) Raw FastQ File --> De Novo Assembly --> Extract All Contigs --> Map Contigs to Reference --> SNPs discovered 1,383

I know from lab data that the consensus sequence produced by method number 2 is closest to the consensus sequence of the bacteria in reality that was sequenced. Does this make method 2 (and it's associated mapped reads) more reliable for SNP calling????

Very keen to receive as many views as possible on this and your reasons as to why/why not.

Many thanks in advance

lg36

[Heisman]

How well worked out is the reference sequence? Do trends in the SNPs that are specific to one method follow trends that are common for all SNPs (ie, similar Ti/Tv ratio for unique calls as for high calls, for example)?