Comparing SNPs amongst populations

[soonerSEQ]

I am new to the next-gen sequencing world (and the human genetics world for that matter), and I am trying to analyze a resequencing data set with some older data.

First, I want to know the number of variants in each population (African American, European American, etc).

Second, I want to know how many and which variants are shared amongst the populations.

Third, I want to know which variants (SNPs), if any, are novel.

I am familiar with command-line programs, but I am so new to this field that I don't even know which programs to use.

I have all my data in .vcf file formats. I have been able to use vcftools to extract all the data for the gene of interest. I tried the --diff command in vcftools, but that didn't give me what I want. This seems like a straight-forward issue but I don't know who to ask.

Thanks for any help.

While I am here, I could also use some recommendations on getting up-to-speed on the topics of LD, haplotypes, and other population statistics. I need a bioinformatics/genetics for dummies.

Thanks!

EDIT:
Just want to say that I have tinkered around with PLINK, vcftools, R, and Haploview.

[JackieBadger]

GATK and samtools are a good place to start

[soonerSEQ]

GATK and samtools are a good place to start

Thanks - I will add those to my list!

Edit:

I was able to print the list of SNPs by using PLINK --write-snplist command. I then compared this list with my other populations to determine which were shared and which weren't (by using a crude method in excel).

Is there not a more scientific way to do this? Is this even a correct method?

[soonerSEQ]

Could anyone else weight in here? I could really use some guidance.

Thanks!