Answers are grounded in this paper and cite their sources.
In 2002 we reported that SOX6, SRY and the SOX9 HMG domain act in pre-mRNA splicing,and that immunodepletion of SOX6 from HeLa nuclear extract impaired splicing of severalsubstrates (Ohe et al., PNAS 2002). Here the author, a co-first author of that study, testsfour predictions of the claim that SOX6 is a general splicing factor using only public data.(1) In mouse epiblast stem cell-derived oligodendrocyte precursor cells (OPCs) andoligodendrocytes (OLs) with CRISPR knockout of Sox6 by two independent guides(GSE181952), editing of Sox6 was confirmed in the RNA-seq reads, yet splicing efficiencyover 21,371–29,140 introns did not decrease. The knockout-minus-control difference inpooled splicing efficiency was −0.03% in OPCs (exact permutation p = 0.661) and +0.09%in OLs (p = 0.190), with no intron reaching FDR < 0.05. The same pipeline applied topladienolide B-treated mouse hepatocyte organoids (GSE249801) detected a 7.6-fold risein pooled intron retention and 4,281 introns with increased retention at FDR < 0.05. (2) InDepMap CRISPR screens (26Q3), 9 of 1,361 cell lines depended on SOX6, and its top co-dependencies did not include spliceosome genes. (3) SOX6 and SRY were not detected inany of 62 human RNA-interactome datasets compiled in RBP2GO 2.0, whereas U1-70Kwas detected in 39. (4) The Human Protein Atlas annotates SOX6 and SOX9 to thenucleoplasm, not to nuclear speckles. SOX6 mRNA in HeLa was 0–2.5 counts per millionacross four public datasets. These data do not support a general requirement for SOX6 inconstitutive splicing in the cell types examined.
No takes yet. Share an insight, caveat, or question.
Kenji Ohe (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: