Designs For Validating Ngs Variants

Question

[jvijai]

I am trying to get a consensus on the best tools used by the community for validation of variants SNVs, Indels that are discovered from NGS (exome or genome). The old fashioned Sanger sequencing is still our current method since we always try to get fast validation of low hanging fruits, which are just a handful. However, when you have larger numbers, there is a limit to the number of variants you can validate by Sanger where it becomes less attractive due to costs. I know some people who then scale up to a MiSeq, HiSeq, PGM path to validate by doing deep resequencing. Other methods such as Fluidigm, Sequenom or RainDance are also mentioned in the literature.

To this end, primer design in batches for Sanger is one of the options. Is there a tool that can use a (hg19) genomic coordinates from VCF or BED to pull sequences, analyze for cross hybridizations, and spit out primers for a moderate number of variants? Specifically, has anyone tweaked primer3 or primer4 to use the above as input to generate an "orderable' list of primers? I like PrimerBLAST.

I know of tools such as BatchPrimer3 which are geared towards plant genomes, since they use those genome as mispriming libraries.Other such as PCRTiler, primers4clades, SCPrimer, are specific for very specific PCR applications.
How do you design your technical validation of NGS variants?

Answer 1

[Pierre Lindenbaum]

For primer3 , one could generate a list of primers from a VCF file using something like:

IFS="\n"
grep -v "#" input.vcf | while read L
do
    C=`echo $L | cut -d '    ' -f 1`
    P=`echo $L | cut -d '    ' -f 2`
    S=$(( $P - 1000))
    E=$(( $P + 1000))
    echo "PRIMER_SEQUENCE_ID=${C}_${P}"
    echo "SEQUENCE_TARGET=990,20"
    echo -n "SEQUENCE_TEMPLATE="
    samtools faidx \
        /path/to/hg19.fa ${C}:${S}-${E} |\
        grep -v ">" | tr -d "\n" | tr "atgcn" "ATGCN" 
    echo
    echo "=" 


done | primer3_core -default_version=1