What's wrong with PLINK?
Dear all
I am giving a course about bioinformatic tools used for genetic association studies. The target audience are phd students with a medical/biological background. Since non of these have experience within programming/linux/R etc I will like to stick to "user friendly" tools like the following:
SNAP - investigate LD, find out whether a SNP is on GWAS arrays
Hapmap, haploview - select tag SNPs
locusZOOM - visualize association plots
UCSC - investigate the properties/positions etc of SNPs
plink - make association studies and QC for GWAS data
NHLBI Exome Sequencing Project (ESP) - Get sequence derived SNP information
I know the list could probably be expanded with 100s of good ideas and the term "user friendly" is very subjected, I know.
But, please fill in. I would very much like to hear from you
All the best
Thomas
6 answers
I would happily add tools/databases like
PolyPhen / SIFT : Effect prediction, even though the percentage of coding SNPs are less, will demonstrate this to show some nice tools in the area of effect prediction.
GWAS Catalog : to show the available repertoire of genotype-phenotype data
HuGENavigator : to introduce how the genetic association data is integrated with relevant information
PheGenI : to introduce how GWAS data is integrated with NCBI resources
DistiLD :Very interesting database of diseases and traits in linkage disequilibrium blocks, can be used to discuss various interesting LD regions associated diseases revealed by GWAS.
We use SAS and HelixTree from GoldenHelix.
Lmekin - a linear mixed model using kinship - is what we use to analyze GWAS data comprised of sets of related individuals. Kinship is often ignored or under-appreciated, but one would do so at one's peril. There are several internet sites with Lmekin info, many in R.
Haploview is preferred for calculation of LD.
We do not like PLINK.
ANNOVAR and VAAST are good for discerning functional consequences of SNP alleles.
Both dbSNP and OMIM have information on the clinical significance of SNPs.
Second @Larry_Parnell on VAAST. Our lab has a mini course on how to use VAAST to find the causal variant in Miller Syndrome.
SWEEP from Broad is also nice for EHH.
Both of these aren't for classic GWAS.
a couple we have used:
Genevar - for discovery and visualization of eQTL data http://www.sanger.ac.uk/resources/software/genevar/
mmimiRNA - a nice tool for inspecting mRNA/microRNA co-expression
mimirna: http://mimirna.centenary.org.au
I would to add following tools/database
PolyPhen / SIFT : Effect prediction, even though the percentage of coding SNPs are less, will demonstrate this to show some nice tools in the area of effect prediction.
GWAS Catalog : to show the available repertoire of genotype-phenotype data
HuGENavigator : to introduce how the genetic association data is integrated with relevant information
PheGenI : to introduce how GWAS data is integrated with NCBI resources
DistiLD :Very interesting database of diseases and traits in linkage disequilibrium blocks, can be used to discuss various interesting LD regions associated diseases revealed by GWAS.
Also check out variant tools varianttools.sourceforge.net), which provides a complete toolset for association studies.
Log in to answer this question.
Thanks for all the nice response. This is indeed a very big help. Thomas