SNP positions in dbSNP and Cosmic databases

[kalinka23]

Hi,

i am doing some SNP calling and i would like to know if the called SNPs are present in the dbSNP and / or in the Cosmic databases. Therefore i have to look at the positions of the called SNPs and then at the positions given in the databases.

But i don't understand exactly how to interprete the database columns. In both, there are two given positions of the variation: here we have for example the first 5 columns of two different dbSNP entries:
585 chr1 10326 10327 rs112750067
585 chr1 10433 10433 rs56289060

and here, the positions of two different Cosmic database entries:
Mutation GRCh37 genome position: 17:7578276-7578277
Mutation GRCh37 genome position: 17:7578433-7578433

When only one position is given then its clear for me (like 10433 10433 or 17:7578433-7578433). But i don't know which position corresponds to the exact position of the SNP in entries with positions like 10326 10327 (the first dbSNP example) or 17:7578276-7578277 (the first cosmic example). Which one of the two positions corresponds to the exact position of the given SNP?

Thank you for any help,

best regards, Karin

[Milan Radovich]

Ill speak for dbSNP only, but those entires that have multiple positions like your example "10326 10327" are indels, where those SNPs that have positions like "10433 10433" are SNPs. dbSNP has both SNPs and indels as does COSMIC. One good software to use to annotate your SNPs is ANNOVAR.

[krobison]

In theory, COSMIC has virtually no SNPs -- it is intended to be somatic mutations only (SNVs). However, some SNPs may have sneaked in due to some tumor samples not having paired germline samples.

[kalinka23]

Thank you very much Milan and krobinson,

i had a look at the annotation af ANNOVAR. They are using the second position (dbSNP) and i think this is good because doing that you only get those which have an SNP and not the indels because only in case of SNPs these positions are the same as the first...

thanks again and regards,
Karin

[kalinka23]

output breakDancer

Hi everybody,

i have a question concerning the output of breakdancer: if i call variants with the commands:

breakdancer-max1.2/bam2cfg.pl -g -h tumor.bam normal.bam > my.cfg
breakdancer_max my.cfg > my.ctx

the called variants are located in the tumor sample, right? I am confused and this is probably a very stupid question, but, in my output, i see for example the following:

chr18 98582 22+1- chrY 11929053 25+2- CTX -214 99 21 normal.bam|21
chr18 98981 2+7- chrY 11928187 1+5- CTX -214 99 5 normal.bam|5

does this mean that the variants are located in the tumor sample based on (or compared to) the normal sample? And where can i find a manual to understand better all the output columns?

Thanks a lot!

Best, Karin