Preprint Genome-wide association studies of Long COVID and post-acute complications of SARS-CoV-2 in the UK Biobank Data, 2025, Prieto-Alhambra et al.

[SNT Gatchaman]

Genome-wide association studies of Long COVID and post-acute complications of SARS-CoV-2 in the UK Biobank Data

Spoiler: Authors

Daniel Prieto-Alhambra; Marta Alcalde-Herraiz; Kim López-Güell; Shahed Iqbal; Jeffrey Wallin; Yunhao Liu; Jun Xie

The genetic foundations of post-COVID-19 conditions remains unclear. We performed two genome-wide association studies (GWAS) in UK Biobank COVID-19 positive individuals to identify the genetic variants associated with Long COVID (LC) and post-acute cardiovascular complications of SARS-CoV-2 (PACS-CVD).

The LC cohort comprised 8,469 participants (68% cases). The PACS-CVD cohort included 105,175 individuals (2% cases). LC GWAS identified 15 independent signals at suggestive significance (p-value<5×10⁻⁶), with 73.3% validated.

The fully validated variant, rs12335232 (ADCY8), has been linked to memory decline, COVID-19 infection and severity. Other loci were near CHRNA7 (neuroinflammation, COVID-19 severity) and RNU7-126P (COVID-19 hospitalization). These findings consistently demonstrate shared biological pathways between acute infection and persistent symptoms. PACS-CVD GWAS identified 14 suggestive loci, mainly near genes linked to cardiovascular and metabolic functions (SAYSD1/KCNK5, FLT1) or COVID-19 severity (ROR2).

These results enhance the genetic understanding of Long COVID and PACS-CVD pathophysiology and highlight several potential therapeutic targets for both conditions.

Web | PDF | Preprint: Research Square | Open Access

 

[SNT Gatchaman]

Although no SNPs reached genome-wide significance (p-value ≤ 5×10−8 ) in the LC GWAS, 15 genomic loci achieved suggestive significance threshold (p-value ≤ 5×10−6)

Of these, one variant -rs12335232 in the ADCY8 gene- was fully validated (Figure 2), with the G allele associated with an increased risk of LC (OR = 1.32, 95%CI = 1.17–1.49).

GeneCards — ADCY8

Zhang et al Dissecting the genetic complexity of myalgic encephalomyelitis/chronic fatigue syndrome via deep learning-powered genome analysis (2025) highlighted ADCY10.

 

[SNT Gatchaman]

Fatigue subtype LC from supplementary table S7 —



RPL5P19
POU6F2
TCF7L2
GRK5
CDON
CALCOCO1
DCAF4
YIPF2

 

[ME/CFS Science Blog]

If I understand correctly, the LC GWAS had only 5,768 cases and 2,701 which is really small so no wonder they didn't find any significant hits.

Here's what their Manhatten plot looks like:

I haven't checked systematically but most of these do not ring a bell in relation to DecodeME. It also seems that many of the suggestive regions were stacked with genes so hard to tell which ones are causally related. They also note poor replication with prior Long Covid studies:

A critical limitation in Long COVID genetics research is the lack of consistent replication across studies49. None of the genetic loci identified inprior studies, including those from our analysis, have achieved genome-wide replication across independent cohorts. For example, even thestrongest FOXP4 association in the HGI did not pass statistical significance in the subsequent 23andMe analysis, despite the latter study havingover eight times more cases