I am currently analyzing a dataset of tumors from different patients (171 patients), with each several mutations (in 170 genes). I want to assess the gene that are mutually exclusive and those who are co-occurent in the dataset. I have the function "discoversomaticInteractions", allowing me to generate a table with pValue, OddsRatio and the pair of gene analyzed, along with a nice visualization of the interactions between my gene.
1. The pValue is not corrected if I am not mistaken. I have put two thresholds like in the vignette, 0.01 & 0.002. I am wondering if those in a publication are acceptable ? Shouldn't I correct it ?
2. Even if the correction doesn't work, what is your opinion on exploiting the results of the "discoversomaticInteractions" function as it is ? Here is an exemple of an output.
3. The "Event" column value is always NA. Can I change it to "co-occurrent" when the OddsRatio is positive, and "mutual_exclusive" when it is negative ?
Hello,
I am currently analyzing a dataset of tumors from different patients (171 patients), with each several mutations (in 170 genes). I want to assess the gene that are mutually exclusive and those who are co-occurent in the dataset. I have the function "discoversomaticInteractions", allowing me to generate a table with pValue, OddsRatio and the pair of gene analyzed, along with a nice visualization of the interactions between my gene.
My table : gene1 gene2 pValue OddsRatio 00 11 01 10 Event pair event_ratio TET2 PRRC2C 0.0008754013 3.057793 128 17 17 9 NA PRRC2C, TET2 17/26 SPEN TP53 0.0021338900 -2.670828 87 9 46 29 NA SPEN, TP53 9/75 NOTCH2 PRRC2C 0.0023754897 2.624247 127 16 18 10 NA NOTCH2, PRRC2C 16/28 EP300 ARID1A 0.0030133890 2.520945 138 11 12 10 NA ARID1A, EP300 11/22 STAT6 KMT2D 0.0035741349 2.446829 114 18 35 4 NA KMT2D, STAT6 18/39 ARID1A TET2 0.0048790823 2.311662 134 12 14 11 NA ARID1A, TET2 12/25
I have several questions :
1. The pValue is not corrected if I am not mistaken. I have put two thresholds like in the vignette, 0.01 & 0.002. I am wondering if those in a publication are acceptable ? Shouldn't I correct it ?
Here is a sum-up of the data :
I tried this command below, but got this error that I have not understand (I checked the structure, don't see any NA or Inf values).
I tried also this but got the exact same corrected value, I guess there are too many events, but I would have expected pValue more
> p.adjust(B_results$pValue, method = "BH") [1] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [14] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [27] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [40] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [53] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [66] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [79] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [92] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [105] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [118] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [131] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [144] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [157] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [170] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [183] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [196] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [209] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [222] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [235] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [248] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [261] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [274] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [287] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [300] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [313] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [326] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 [339] 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2430722 0.2441138 0.2447317 [352] 0.2447317 0.2447317 0.2447317 0.2448216 0.2448216 0.2448216 0.2448216 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 [365] 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 [378] 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 [391] 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 0.2451427 [404] 0.2451427 0.2451427 0.2451427 0.2455163 0.2455163 0.2455163 0.2455163 0.2455163 0.2455163 0.2455163 0.2455163 0.2455163 0.2485581 [417] 0.2485581 0.2485581 0.2488121 0.2488121 0.2488121 0.2488121 0.2489317 0.2496912 0.2496912 0.2496912 0.2496912 0.2496912 0.2496912 [430] 0.2496912 0.2496912 0.2496912 0.2496912 0.2496912 0.2496912
2. Even if the correction doesn't work, what is your opinion on exploiting the results of the "discoversomaticInteractions" function as it is ? Here is an exemple of an output.
3. The "Event" column value is always NA. Can I change it to "co-occurrent" when the OddsRatio is positive, and "mutual_exclusive" when it is negative ?
Complete code :