SOAPsnp - SNP information file

[dmb]

SOAPsnp is able to take a SNP information file for use in calling SNPs. Unfortunately, the format of this file seems to be specific to SOAPsnp. I can pull data from NCBI's dbSNP archive but this file is in a different format AND does not contain all the information that SOAPsnp takes as input. I can parse the dbSNP files (one for each chromosome) and write to the SOAPsnp format but if this can be pulled from somewhere I don't know of, that would be preferred.

Does anyone use the -s option when using SOAPsnp?
If you use it, did you have to generate your own SNP info file?

Also, it looks like some of the attributes that SOAPsnp takes can be "zeroed" out if the data is unknown - e.g.
frequency of A,
frequency of C,
frequency of T,
frequency of G
...or given a default value if unknown - e.g.
whether the SNP is validated by experiment (1 is true, 0 is false),
whether the SNP is actually an indel (1 is true, 0 is false).

This is not the method I would prefer but I guess its better than not using the -s option at all.

Thanks,
Dan

[dmb]

follow up

To those that might be interested...

I found somewhat of a solution. I used the Table Browser from UCSC to extract SNP data using the following parameters.
group: All Tables
database: hg19
table: snp131
region: genome
I output this to a file (tab-delim) that has the following header information:
#bin
chrom
chromStart
chromEnd
name
score
strand
refNCBI
refUCSC
observed
molType
classvalid
avHet
avHetSE
func
locType
weight

One of the rows of data looks like this:
585 chr1 10491 10492 rs55998931 0 + C C C/T genomic single unknown 0 0 near-gene-5 exact 1

Using some simple R code I should be able to create the SOAPsnp SNP info file.

If anyone has another method, I would love to hear about it.

[dmb]

the file from UCSC has 26,033,054 rows.

[asangqi]

As far as I know (form my experience)
the format of dbsnp file which soapsnp needs is:
chromosome name: should same with reference chr name
position:
whether including allele frequency: (0/1) 0:no 1:yes
whether this mark be validate: (0/1)
is this mark is a indel?: (0/1)
frequency of A,
frequency of C,
frequency of T,
frequency of G
snp ‘s number

so you’d better not use “-“ , the SoapSNP may cannot realize it( the bug of soapsnp, may be ,I guess)

[biomaseq]

SOAPsnp is able to take a SNP information file for use in calling SNPs. Unfortunately, the format of this file seems to be specific to SOAPsnp. I can pull data from NCBI's dbSNP archive but this file is in a different format AND does not contain all the information that SOAPsnp takes as input. I can parse the dbSNP files (one for each chromosome) and write to the SOAPsnp format but if this can be pulled from somewhere I don't know of, that would be preferred.

Does anyone use the -s option when using SOAPsnp?
If you use it, did you have to generate your own SNP info file?

Also, it looks like some of the attributes that SOAPsnp takes can be "zeroed" out if the data is unknown - e.g.
frequency of A,
frequency of C,
frequency of T,
frequency of G
...or given a default value if unknown - e.g.
whether the SNP is validated by experiment (1 is true, 0 is false),
whether the SNP is actually an indel (1 is true, 0 is false).

This is not the method I would prefer but I guess its better than not using the -s option at all.

Thanks,
Dan

I have the same promoble!