Rank aggregate calculation

[adzikjurasz]

Hi! Thank you for access to this very helpful and educational piece! I managed to follow and replicate the individual calculations i.e. PRR, GPS, but came unstuck during the rank aggregate calculation. I managed to find the files"df_ror, df_bcpnn and df_prr but not df_gps from my earlier individual calculations (i.e. PRR module") I had a guess at which file df_gps might be, but whatever I try I get:

df_gps <- readRDS("data-processed/df_gps.rds")
Error in gzfile(file, "rb") : cannot open the connection
In addition: Warning message:
In gzfile(file, "rb") :
cannot open compressed file 'data-processed/df_gps.rds', probable reason 'No such file or directory'

I believe my $PATH is correct otherwise I wouldn't have gotten this far

Even if I "ignore" reference to df_gps, I get:

rownames(df_prr) <- paste(df_prr$drug code, df_prr$event effect, sep = " <> ")
rownames(df_ror) <- paste(df_ror$drug code, df_ror$event effect, sep = " <> ")
rownames(df_bcpnn) <- paste(df_bcpnn$drug code, df_bcpnn$event effect, sep = " <> ")
common_signals <- Reduce(intersect, list(rownames(df_gps), rownames(df_prr), rownames(df_ror), rownames(df_bcpnn)))
length(common_signals)
[1] 0

If you could help I would be most grateful

Andy J

[nanxstats]

@adzikjurasz By knitting analysis-gps.Rmd it will save the output data frame as data-processed/df_gps.rds file in the end. I just clone the repo to run it again and got the identical results as before (a 311 x 7 data frame). So probably check if there are any issues when knitting this file. Running through it could take quite a few minutes.