Comparative Snp Analysis

Question

[GPR]

Hello, I am trying to compare the degree of A-to-G editing in a near-to-isogenic pair of cell lines. I have two biological replicates and have mapped with Bowtie and BWA, followed by a samtools mpileup | VarScan analysis. After this, I have used bedtools intersect to extract variants not annotated in dbSNP, but are in Alu repeats. Here is where I have some doubts, mainly two questions: QUESTION 1: In the vcf file (VarScan output),

#CHROM  POS     ID      REF     ALT     QUAL    FILTER    INFO    FORMAT  Sample1    Sample2
   chrM    73      .           G       A       PASS     DP=238  GT:GQ:DP           1/1:71:121  1/1:69:117

What exactly is the meaning of

FORMAT   Sample1    Sample2
GT:GQ:DP 1/1:71:121  1/1:69:117

QUESTION 2:

I have higher number of editing sites "called" in sample 1 than in sample 2 in the 1st biological replicate (about 16% difference). However this difference is reversed in the 2nd biological replicate. What is the proper way of comparing the degree of RNA editing in two different samples? Is there a quantitative procedure? I have naively compared them with bedtools intersect, using or omitting option -v. Is this the correct way to go about it?

Many thanks. G.

Answer 1

[zx8754]

Regarding Question 1 see Genotype fields at http://www.1000genomes.org/wiki/Analysis/Variant%20Call%20Format/vcf-variant-call-format-version-40

[GPR]

Thanks so much!

Answer 2

[dankoboldt]

Thanks for answering, zx8754! And GPR, if you have any VarScan-related questions, please let me know.