Open scisea opened 4 years ago
bunop
commented
4 years ago Dear @scisea ,
Thank you for your interest in detectRUNs!
for the "count and count_chr" tables it would be very useful to obtain the mean ROH values ( plus standard dev/error) according to the number of samples per population.
This seems reasonable to me. @filippob , what do you think? could this feature be added in the next release?
filippob
commented
4 years ago Yes, we are taking note of detected bugs and proposed enhancements for the next release. We received a number of suggestions on how to improve the stats part of the package, and will certainly consider these. Thank you!
On Wed, Jun 3, 2020 at 9:41 AM Paolo Cozzi notifications@github.com wrote:
Dear @scisea https://github.com/scisea , Thank you for your interest in detectRUNs!
for the "count and count_chr" tables it would be very useful to obtain the mean ROH values ( plus standard dev/error) according to the number of samples per population.
This seems reasonable to me. @filippob https://github.com/filippob , what do you think? could this feature be added in the next release?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/bioinformatics-ptp/detectRUNS/issues/16#issuecomment-638020430, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABGARJHCYCRVJEQDW63UWYLRUX5BFANCNFSM4NQI35AA .
Filippo Biscarini PhD, Researcher
Bioinformatics, biostatistics, genetics (Milan, Bruxelles, Lodi)
The data obtained from the summaryRuns option is great, however, for the "count and count_chr" tables it would be very useful to obtain the mean ROH values ( plus standard dev/error) according to the number of samples per population. That way we can really compare them. Right now we get a total number which could be biased when the pops size is different. Same for the ROH length cause we get the mean but not the standard deviation of the values.
I think it would be straightforward to achieve for someone with good R skills not like me hah but I think that by adding these options to the ddply function "SE = sd(N) / sqrt((length(N))), SD = sd(N))" ??
Anyway it is just a suggestion but I think it would be very much appreciated. Cheers!