EpiSignaLR: an open resource for genome-wide episignature analysis from methylation-aware short- and long-read sequencing

Differentially methylated regions identified from Oxford Nanopore data, classifiers built with nested cross-validation

Episignatures, disorder-specific patterns of genome-wide DNA methylation established for more than 100 Mendelian conditions, are currently tested mostly with methylation arrays, which are limited to their probe set. The authors sequenced individuals with SETD1B-related neurodevelopmental disorder, Kabuki syndrome or Kleefstra syndrome, along with controls, using Oxford Nanopore, PacBio HiFi or Illumina 5-base whole-genome sequencing. They identified 143, 20 and 23 differentially methylated regions for the three signatures directly from ONT data, 15 to 26% of which contained no array probe. Cross-validated classifiers ranked every case above every control; the final classifiers called all but one held-out case correctly, the remaining case carrying a KMT2D missense variant being indeterminate, and the SETD1B classifier correctly categorized seven of seven cases at two outside sites. EpiSignaLR is an open workflow combining these three ONT-derived classifiers with 34 published array-derived signatures.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

Preprint, not peer reviewed. Folding episignatures into long-read sequencing means obtaining, in a single test, both the variants and a functional argument for reclassifying a variant of uncertain significance, without a separate methylation array. The authors acknowledge the limitation: cohorts too small per condition and one PacBio control falling towards the Kabuki cases, a sign of platform-specific methylation differences to be corrected before any diagnostic use outside ONT.

Analysis by Dr Thibaut Benquey