Problem with Manhattan plot

Question

[lurdesdl]

Hi all,

I have run a linear regression association test on PLINK, and then I have represented the results with a Manhattan plot. The problem is that a strange horizontal alignment appears in the representation (around Pvalue = 10^(-7)). I tried to run a second association test with another phenotype, however this horizontal "line" still persists (this time at the top of the graphic). It seems to be an artefact, but I can't identify the cause. What could be the problem?

[TriS]

how do your data look? what's the code that you used? I suppose you are using R for the plot? what does the image look like (we can't see it, or at least I can't)? hard to help without info

[lurdesdl]

1) Here is a link where you can download the dataData link. There are two files: Res1.assoc.linear and Res2.assoc.linear, corresponding respectively to ManhattanPlot1 and ManhattanPlot2

2,3) Yes, I'm using R for the Manhattan plot. The code is very simple:

#Required packages installation
if (!require("data.table")) install.packages("data.table")
if (!require("qqman")) install.packages("qqman")

#Setting directory and the corresponding file
setwd('...')
file <- 'Res1.assoc.linear'

#Reading data
data <- fread(file, header = TRUE)
#Filtering NA P-values
data_F = subset(data, is.na(data$P)!=TRUE)

#Plot
manhattan(data_F)

4) To see images do right-click on the icon and select "Open in New Tab"

[SGMS]

Hi lurdesdl,

I was wondering whether you managed to come up with a solution to your problem above. It seems I also have a a line at different p-value thresholds e.g 10^-7, 10^-5 etc. It would be great if you could give me a hint.

Thanks a lot

[abc]

Have you had a look at this similar question? Worked for me - it seems to be related to the MAF.

[SGMS]

Thanks, I have seen it and it also helped me!