Two minor bugs

[philippbayer]

Hi

Thank you for the recent improvements to this package!

I have two minor things:

1. When rMVP reports the significance level, it cuts off at 6 decimal places:

   rMVP/R/MVP.r

   Line 307 in 3acd666

   logging.log(paste("Significant level: ", sprintf("%.6f", threshold/m), sep=""), "\n", verbose = verbose)

   
   In my case I have so many SNPs that rMVP always prints Significant level: 0.000000. Perhaps it would be better to use scientific notation using formatC(0.0000000000005, format = "e", digits = 2)?

2. The files ending in signal.csv are very useful, is it possible there is a bug when only a single SNP is significant (which happens a lot with FarmCPU)?
   Normally, I get a proper table with proper headers, but when I have only one SNP, I get a strange text-file like this:

"x"
"SNP_5015"
"3"
"47924314"
"-6185.31927432418"
"1015.52184827578"
"1.50018111376143e-08"

i.e., the single-row table got transposed. Funny enough it's just a cbind and a write.csv, I currently don't understand how this happens.

sessionInfo():

R version 3.6.1 (2019-07-05)
Platform: x86_64-conda_cos6-linux-gnu (64-bit)
Running under: Red Hat Enterprise Linux Server release 6.10 (Santiago)

locale:
 [1] LC_CTYPE=en_AU.UTF-8       LC_NUMERIC=C
 [3] LC_TIME=en_AU.UTF-8        LC_COLLATE=en_AU.UTF-8
 [5] LC_MONETARY=en_AU.UTF-8    LC_MESSAGES=en_AU.UTF-8
 [7] LC_PAPER=en_AU.UTF-8       LC_NAME=C
 [9] LC_ADDRESS=C               LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_AU.UTF-8 LC_IDENTIFICATION=C

attached base packages:
[1] parallel  stats     graphics  grDevices utils     datasets  methods
[8] base

other attached packages:
[1] rMVP_0.99.16     bigmemory_4.5.33 MASS_7.3-51.4

loaded via a namespace (and not attached):
[1] bigmemory.sri_0.1.3 compiler_3.6.1      tools_3.6.1
[4] Rcpp_1.0.2

[XiaoleiLiuBio]

Hi Philipp Bayer, Thank you very much for your nice suggestions. We will solve the problems as you suggested. Cheers, Xiaolei

----------------

---------- Origin message ----------

From："Philipp Bayer" ***@***.***> To："xiaolei-lab/rMVP" ***@***.***> Subject：[xiaolei-lab/rMVP] Two minor bugs (#44) Date：2019-10-28 11:42:28Hi

Thank you for the recent improvements to this package! I have two minor things: 1) When rMVP reports the significance level, it cuts off at 6 decimal places: https://github.com/xiaolei-lab/rMVP/blob/3acd666df3236da4ee40c52ffe30e8f45914eca0/R/MVP.r#L307 In my case I have so many SNPs that rMVP always prints `Significant level: 0.000000`. Perhaps it would be better to use scientific notation using `formatC(0.0000000000005, format = "e", digits = 2)`? 2) The files ending in signal.csv are very useful, is it possible there is a bug when only a single SNP is significant (which happens a lot with FarmCPU)? Normally, I get a proper table with proper headers, but when I have only one SNP, I get a strange text-file like this: <pre> "x" "SNP_5015" "3" "47924314" "-6185.31927432418" "1015.52184827578" "1.50018111376143e-08" </pre> i.e., the single-row table got transposed. Funny enough it's just a cbind and a write.csv, I currently don't understand how this happens. sessionInfo(): <pre> R version 3.6.1 (2019-07-05) Platform: x86_64-conda_cos6-linux-gnu (64-bit) Running under: Red Hat Enterprise Linux Server release 6.10 (Santiago) locale: [1] LC_CTYPE=en_AU.UTF-8 LC_NUMERIC=C [3] LC_TIME=en_AU.UTF-8 LC_COLLATE=en_AU.UTF-8 [5] LC_MONETARY=en_AU.UTF-8 LC_MESSAGES=en_AU.UTF-8 [7] LC_PAPER=en_AU.UTF-8 LC_NAME=C [9] LC_ADDRESS=C LC_TELEPHONE=C [11] LC_MEASUREMENT=en_AU.UTF-8 LC_IDENTIFICATION=C attached base packages: [1] parallel stats graphics grDevices utils datasets methods [8] base other attached packages: [1] rMVP_0.99.16 bigmemory_4.5.33 MASS_7.3-51.4 loaded via a namespace (and not attached): [1] bigmemory.sri_0.1.3 compiler_3.6.1 tools_3.6.1 [4] Rcpp_1.0.2 </pre>

…

-- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: #44

[philippbayer]

Ah, cbind.data.frame fixed it, that explained why my minimal example using data-frames only didn't replicate that issue, thanks for the fast turnover :)