No significant --clump results. Skipping.

[James Pirruccello]

I am clumping SNPs in a GWAS with fairly weak signal with only a few genomewide-significant hits - many of which are likely to be artifact due to low MAF. However, I usually clump first and then MAF-filter later. It's therefore quite possible that clumping should find that all of my SNPs are independent. However, if I understand the output, plink does not produce output when clumping yields no significant results (hence "skipping"). In terms of stitching together a consistent pipeline, this is not desirable - it would be nice for this clumped file to be produced even if no SNPs are clumped together and all are independent. My main question is whether my understanding of what plink is doing here is accurate. My secondary question, if so, is whether there is a toggle to allow the clump file to be produced even under these circumstances.

Output is below.

Thanks!

- James

PLINK v1.90b6.10 64-bit (17 Jun 2019) www.cog-genomics.org/plink/1.9/ (C) 2005-2019 Shaun Purcell, Christopher Chang GNU General Public License v3 Logging to /mnt/data/snps.gwas.tsv-plink.log. Options in effect: --bfile subsetted --clump /mnt/data/snps.gwas.tsv --clump-field P_BOLT_LMM --clump-kb 5000 --clump-p1 5e-8 --clump-p2 5e-8 --clump-r2 0.001 --clump-snp-field SNP --double-id --make-founders first --out /mnt/data/snps.gwas.tsv-plink --threads 16 64326 MB RAM detected; reserving 32163 MB for main workspace. 66 variants loaded from .bim file. Note: Skipping --make-founders since there are no nonfounders. 36519 people (0 males, 0 females, 36519 ambiguous) loaded from .fam. Ambiguous sex IDs written to /mnt/data/snps.gwas.tsv-plink.nosex . Using 1 thread (no multithreaded calculations invoked). Before main variant filters, 36519 founders and 0 nonfounders present. Calculating allele frequencies... 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% 10% 11% 12% 13% 14% 15% 16% 17% 18% 19% 20% 21% 22% 23% 24% 25% 26% 27% 28% 29% 30% 31% 32% 33% 34% 35% 36% 37% 38% 39% 40% 41% 42% 43% 44% 45% 46% 47% 48% 49% 50% 51% 52% 53% 54% 55% 56% 57% 58% 59% 60% 61% 62% 63% 64% 65% 66% 67% 68% 69% 70% 71% 72% 73% 74% 75% 76% 77% 78% 79% 80% 81% 82% 83% 84% 85% 86% 87% 88% 89% 90% 91% 92% 93% 94% 95% 96% 97% 98% 99%Warning: No significant --clump results. Skipping. done.

[James Pirruccello]

Actually, revisiting the data, I think in this case there are 0 SNPs that achieve the specified P threshold. From a pipeline standpoint, it would still be nice to have a file with the header and no SNPs, but I can catch this on my end as well if that's not an option.

Thanks,

James

[Christopher Chang]

Correct, every index variant in --clump's output passes the --clump-p1 threshold, so if there are no variants satisfying that condition, --clump's output is empty by design.  With that said, I can see why it'd be useful for scripting purposes to generate a file with a header line rather than no file at all here; I'll try to remember to include a way to do that when --clump is added to plink 2.0.