Creating instrumental variable for mendelian randomisation - in Ubuntu

[rmgpanw]

Problem summary

• Hi, I'm trying to create an instrumental variable for age-related macular degeneration (AMD). I have downloaded summary data for the following study https://www.ebi.ac.uk/gwas/studies/GCST003219
• Sample of data shown below:

Marker  Chrom   Pos Allele1 Allele2 Ncases  Ncontrols   GC.Pvalue   Overall
rs12083781  1   796375  C   T   16144   17832   0.0829  +
rs75932129  1   796767  A   G   16144   17832   0.101   -
rs58013264  1   797440  C   T   16144   17832   0.0836  +
rs10900604  1   798400  G   A   16144   17832   0.0967  +
rs11240777  1   798959  A   G   16144   17832   0.102   +
rs61768212  1   801467  C   G   16144   17832   0.0846  +
rs7553096   1   802026  A   G   16144   17832   0.0787  +
rs10157494  1   802496  T   C   16144   17832   0.103   +
rs7526310   1   804759  T   C   16144   17832   0.0799  +

or download here: sample.txt

• I would like to filter filter for only those SNPs with a 'GC.Pvalue' of significance p < 0.00001

I have tried the following, which does not work

awk -F '{if ( $8 = 0.0000.) print $0 }' sample.txt

I get 'syntax error' when I run this with the sample.txt file

Could someone please advise me on how to improve on this? Thanks!

[alhenry]

I think there are a couple of things that caused syntax errors in the initial code:

• -F (field separator) option needs to be followed by an argument. However, I think this can be omitted as the default is whitespace so tab / space should be fine.
• = operator should be < (or == for "equal to", but that's not what we want)
• 0.0000. should be 0.00001

The following should work: awk '$8 < 0.00001' sample.txt

(it wont show any output as there is no variant with P < 0.00001, so to check you can change it to e.g. awk '$8 < 0.1' sample.txt

If you want to preserve the header: awk 'NR==1 || $8 < 0.1' sample.txt