This video gives an overview of ways how to narrow down your SNP genotype data to the region you are most interested in. It also shows examples of output files.
I want to do Genome wide association study(GWAs) and I’m using single nucleotide polymorphism (SNPs) I have 173 snp data generated from our lab and I still need to add more up to 500 the remaining we can get them form online 3k rice data base how to extract those one form the 3000 data using Linux command
Hi, the GWAS is one of the things I want to follow up on in the near future. As for your question, the answer depends on which format the 3K data is in. If it is a PLINK format (ped, bunary ped,vcf, or any other format PLINK is able to read), you can use the --extract option of PLINK, while specifying the exact SNP names you want to get out. If it is some kind of non standard format, you have to carefully extract the rows or columns corresponding to your chosen SNPs.
Personally I used fitted PCA components in a linear model, but I am not too convinced about these. There seems to be always something remaining via the QQ plot... My current go-to solution is the GEMMA software, which estimates relatedness matrix from genotypes. This is double win, as the software itself uses PLINK input files + the relatedness matrix could be implemented as a population structure/stratification in the follow up step right away bioinformaticshome.com/tools/gwas/descriptions/GEMMA.html
Thank you Dr Gabor, the video is very helpful, l have a question: how can we make a loop in R to extract the snps from different chromosomes say 22 chromosome of human in one step ?
Hi, I am lagging with the answer her and you proabably found out already, but you have to put the PLINK command into an R loop, where you change the --ch option with each iteration as well as the --out file name, so you save the outputs under a different name.
would you please make a video about how to analize genomic data and estimate genomic breeding value (e. g. with GCTA or other practical software)? Thank you in advance. and happy new year
Yes, the GCTA is on the plan, at some point. The estimation of genomic breeding values might also come, at some point, but this needs a substantial preparation time.
Yes. Use the from-kb/to-kb or similar options, possibly combining with --chr, if you want just one specific chromosome. www.cog-genomics.org/plink/1.9/filter#mrange_pos
Thank you Professor for all your videos in Genomics but for the beginners if add some packages from the beginning of downloading and installing plink please
Hi, I am glad you like it! The one-liners are most likely just in the videos, but I put the larger scripts on a pastebin account here: pastebin.com/u/GenomicsBootCamp If any larger script is missing, just let me know and I amend that as well.
In this case I just manualy copy-pasted a few SNPs for the example and to demonstrate the format what is expected. So no specific reason for exactly these in the video. There is usually a good reason why someone want to extract or exclude specific SNPs, so the names of these should appear in a single column.
Thanks for your continuous videos on Genomics. I am also interested to learn the estimation of genetic parameters.
I want to do Genome wide association study(GWAs) and I’m using single nucleotide polymorphism (SNPs) I have 173 snp data generated from our lab and I still need to add more up to 500 the remaining we can get them form online 3k rice data base how to extract those one form the 3000 data using Linux command
Hi, the GWAS is one of the things I want to follow up on in the near future.
As for your question, the answer depends on which format the 3K data is in. If it is a PLINK format (ped, bunary ped,vcf, or any other format PLINK is able to read), you can use the --extract option of PLINK, while specifying the exact SNP names you want to get out.
If it is some kind of non standard format, you have to carefully extract the rows or columns corresponding to your chosen SNPs.
@@GenomicsBootCamp it’s a PLINK format. I will try to see.. thank you Dr..
Thank you. That is very helpful. Please add a link to your scripts .
Thank you very much, your videos are really helpful. I was wondering about the way of controlling population stratification in the association study.
Personally I used fitted PCA components in a linear model, but I am not too convinced about these. There seems to be always something remaining via the QQ plot... My current go-to solution is the GEMMA software, which estimates relatedness matrix from genotypes. This is double win, as the software itself uses PLINK input files + the relatedness matrix could be implemented as a population structure/stratification in the follow up step right away
bioinformaticshome.com/tools/gwas/descriptions/GEMMA.html
@@GenomicsBootCamp Thank you, Prof. Gabor. I will have a look at it. My problem was with the QQ plot, hence, I got a high lambda value
your videos on plink are really good
Thank you very much for the simple and complete explanation.
Thanks for this clear and complete explanation.
Thank you Dr Gabor, the video is very helpful, l have a question: how can we make a loop in R to extract the snps from different chromosomes say 22 chromosome of human in one step ?
Hi,
I am lagging with the answer her and you proabably found out already, but you have to put the PLINK command into an R loop, where you change the --ch option with each iteration as well as the --out file name, so you save the outputs under a different name.
would you please make a video about how to analize genomic data and estimate genomic breeding value (e. g. with GCTA or other practical software)?
Thank you in advance. and happy new year
Yes, the GCTA is on the plan, at some point. The estimation of genomic breeding values might also come, at some point, but this needs a substantial preparation time.
Hi, is it possible to extract not by SNPs variants but regions by giving chr position?
Yes. Use the from-kb/to-kb or similar options, possibly combining with --chr, if you want just one specific chromosome.
www.cog-genomics.org/plink/1.9/filter#mrange_pos
Thank you Professor for all your videos in Genomics but for the beginners if add some packages from the beginning of downloading and installing plink please
Hi Edossa! I am not sure if I understand your request. Could you specify?
@@GenomicsBootCamp how to download and install plink1.90
Dear Gabor! Thank you very much. All your videos are really helpful. Could you tell me where I can found your R acripts?
Hi, I am glad you like it! The one-liners are most likely just in the videos, but I put the larger scripts on a pastebin account here: pastebin.com/u/GenomicsBootCamp
If any larger script is missing, just let me know and I amend that as well.
@@GenomicsBootCamp Got it. Thanks!
Thankyou professor for your tutorial.
I am wondering how you got SNP.txt file.
Esp these few because in my case, am getting snps from chr. 1 to 29.
In this case I just manualy copy-pasted a few SNPs for the example and to demonstrate the format what is expected. So no specific reason for exactly these in the video.
There is usually a good reason why someone want to extract or exclude specific SNPs, so the names of these should appear in a single column.