Page 1 of 1

p values extracted from pacoxph output

Posted: Tue Feb 14, 2017 7:17 am
by shirleydaidai
I used pacoxph function to do survival analysis for my imputation data.
The first few lines of "regression_add.out.txt":

name A1 A2 Freq1 MAF Quality Rsq n Mean_predictor_allele chrom beta_SNP_add sebeta_SNP_add chi2_SNP
rs9628072 G A 0.095 1 1 1 220 0.0954545 22 -0.112741 0.289786 0.15541
rs77961652 G A 0.048 0.96 0.96 0.698 220 0.0479705 22 -0.622749 0.537806 1.54305
rs2071890 G A 0.525 1 1 0.999 220 0.52485 22 -0.261426 0.161514 2.62901

which column is the P value that I can use for further analysis?

Re: p values extracted from pacoxph output

Posted: Wed Feb 15, 2017 11:04 am
by lckarssen
Currently, the ProbABEL output only lists the chi^2 value (calculated via the likelihood ratio test) in the last column of the output. We are planning to add p-value support to the output as well, and work is on the way, but since I don't have direct funding for ProbABEL development anymore, this is low on my priority list, so I can't promise a delivery date. Of course, if you have some funds and/or would like to help out with finishing the p-value work, feel free to contact me so we can discuss the various options.

In the mean time, I think the easiest way is to calculate the p-values yourself based on the chi^2 column (e.g. in R).