SNP and Phenotype Correlation
I have SNP data of two samples, where one is susceptible and another is Resistance. How I correlate some SNPs which is related to Resistance ? Is there any possible statistical method?
• 1,236 views
•
link
updated
by
dthorbur
•
written
by
chimerajit •
•
written
by
chimerajit •
0 answers
No answers yet.
Log in to answer this question.
More posts like this
-
All VS All contamination check for tumor samples
written by mps •
-
What I can do with outliers that I got from ddRAD data?
written by mahnazkiani
-
Stacks program output
written by mahnazkiani
-
Comparing two RNA-seq datasets of same species and treatments
written by sakuraazalea •
-
how to determine the gene involved in antibiotic resistance?
written by sharmilla_487 •
-
SNPs mining using freebayes: problem in using multiple bam file of different ploidy
written by ram.mainali •
-
SNP data analysis
written by caspase8mach •
-
How to tell bloodtype from SNPs data?
written by O.rka
-
Genotyping Platform For Fine-Mapping A Region
written by Stephen
-
Comparing Snps Across Populations
written by Andrea_Bio •
You're going to need more than 1 sample per phenotype to infer any stats.
GWAS, which map physical characteristics to genomic loci, typically include hundreds to thousands of samples.
At minimum, I would expect 3 pooled-seq samples per phenotype for a simple and cheap exploratory experiment. Then you would identify variants that fully segregate by phenotype and identify overlapping gene or regulatory elements, but this is still a pretty underpowered methodology.
Thanks for your response. If I have 3 pulled sample can I go for GWAS? or there is any other simpler statistical methods available?
GWAS was just an example of phenotype mapping methodology and the sample sizes required to use it successfully.
Given you're looking at resistance vs susceptible phenotypes, I still don't think 3 samples is enough. Especially since you provide no information about these samples - are they single individuals or pooled-seq, is this 3 total or per phenotype, and do they have good coverage? These questions are the kind of thing you should be asking yourself/supervisor when designing experiments and/or curating datasets.
If possible, it would be good if you knew more about the trait - is there evidence to suggest it's polygenic, any idea of the penetrance, could it be epistatic, etc...? Assuming full penetrance of a single locus underlying resistance (which is a big assumption), you could identifying candidate variants that segregate by phenotype, but it's likely this list will be large and without further information it would be very hard to whittle that list down to a reasonable dataset for functional characterisation.