Focusing in on truncated genes without any known disease-association, we found that GCG, PDC, and ENTPD3 are compelling candidates as recessive loss-of-function rare disease genes! Shout out to our wonderful collaborators across other cohorts who helped find other probands with GCG/ENTPD3 LoF.
amcguigan.bsky.social
Headlines:
-Nearly 7% of the human genome can be lost without obvious consequence
-One in 200 rare disease patients in the NGRL have a clinically relevant homozygous deletion
-Promoter deletions are a recurrent non-coding cause of disease
-We propose PDC, GCG and ENTPD3 as novel rare disease genes
A huge thank you to everyone involved, including all our wonderful collaborators, the team at Genomics England, and the patients and their families, without which none of this research would have been possible. @nickywhiffin.bsky.social
I’m delighted to share our work leveraging homozygous deletions to discover rare disease diagnoses and novel disease genes - see the preprint here: www.medrxiv.org/content/10.6...
We mapped biallelic deletions across the NGRL cohort and found that 213 million unique nucleotides, corresponding to 6.92% of the genome, were deleted across the cohort. Deletion tolerance varied by annotation; 2.99% of unique MANE transcript CDS sequence was deleted compared to 9.41% of intergenic.
We didn’t stop there! Analysing 5’ and 3’-UTR deletions, we noticed that the most compelling pathogenic variant candidates were in 5’-UTR which also knocked out upstream sequences. Hypothesising that promoter loss may be driving the mechanism, we discovered 19 candidate diagnoses, some with RNA seq
amcguigan.bsky.social
amcguigan.bsky.social
Looking at where these deletions overlapped known disease genes, we found 276 cases where a known disease-associated gene had a CDS truncation arising from a homozygous deletion. This included recurring variants, such as STRC and NPHP1, alongside many unique events.
Structural variants (SVs) can disrupt gene function and contribute to pathogenesis of rare disorders. Here, we created a genome-wide knockout dataset across 125,730 individuals with genome sequencing ...
