You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am a master's student studying plants and VCF format is a commonly used SNP format for me. I use this command to processing my data :
After the phenotype is added, an additional file of the phenotype is output. The output phenotype file is not in the same order as my original phenotype file. After I checked it, it seemed that in subsequent GWAS analyses, it seemed that the output phenotype file had to be used instead of my original phenotype file. But this is not emphasized in your ReadMe. I would like to ask what is the reason for this?
And by the way, in the given command, the row that adds the phenotype is commented out, but shouldn't the line that adds the phenotype be important?
Thank you very much for answering my questions!!!
The text was updated successfully, but these errors were encountered:
Thank you for the question.
Yes, if the phenotype file is provided, rMVP will automatically re-order the individuals in phenotype file to be consistent with that in genotype file, we recommend providing phenotype file since it can help users avoid some potential problems. However, it's not mandatory, because some users are only interested in parts of functions in rMVP, e.g., genotype numeric coding, PCA and Kinship calculation, in which the phenotype file is not involved. Therefore we command out this line in the tutorial script to tell the users that the phenotype file is optional, it does not represent unimportance.
Thank you for your answer
The reason I asked this question was because in the subsequent GWAS analysis, the phenotype file I provided is the original one, not a reorganized version, leading to consistent issues with the results. I think you can mention this in the README of GWAS analysis, maybe it will be more comprehensive!
I am a master's student studying plants and VCF format is a commonly used SNP format for me. I use this command to processing my data :
After the phenotype is added, an additional file of the phenotype is output. The output phenotype file is not in the same order as my original phenotype file. After I checked it, it seemed that in subsequent GWAS analyses, it seemed that the output phenotype file had to be used instead of my original phenotype file. But this is not emphasized in your ReadMe. I would like to ask what is the reason for this?
And by the way, in the given command, the row that adds the phenotype is commented out, but shouldn't the line that adds the phenotype be important?
Thank you very much for answering my questions!!!
The text was updated successfully, but these errors were encountered: