Two thirds of our genome consists of repetitive sequences, and until recently many scientists didn’t care all that much. But as our understanding of the genome improves, so does our insight in the importance of many of these repeats.
One particular tandem repeat in the subtelomeric region of chromosome 4q can cause havoc when the number of repeat units drops. Contraction of a D4Z4 macrosatellite repeat in that region is associated with facioscapulohumeral muscular dystrophy or FSHD, one of the most common hereditary forms of muscle disease.
When that repeat contracts to less than ten copies, a typically repressed transcription factor turns on which triggers the development of the disease. Further complicating the matter, there is an almost identical repeat subtelomeric of chromosome 10q, which is not involved in the disease. And there are two haplotypes of the chromosome 4 region, one of which is pathogenic and one that is not.
In order to genotype FSHD correctly, you must distinguish between chromosome 4 and 10, as well as the pathogenic and non-pathogenic allele, and get an exact copy number count of that repeat.
This week, scientists from Wenzhou Medical University, Wenzhou Central Hospital, Berry Genomics and The First Hospital of Kunming, all in China, published a paper in Molecular Genetics & Genomic Medicine in which Bionano genome mapping correctly characterized the molecular structure of the FSHD locus in the affected individuals of a five-generation pedigree with FSHD1. They identified the founder 4qA disease allele as having four D4Z4 repeat units, and identified a structural variant of the FSHD1 region involving a duplication of a 4qB allele, which appears clinically as a milder form of FSHD1.
They praise Bionano’s “moderate sample requirements and short time frame compared to Southern hybridization” its “potential to identify structural variants such as deletions, duplications, or rearrangements of the locus”, and believe that Bionano “represents an advance in both FSHD1 diagnosis confirmation and has future practical potential in the modern molecular diagnostic laboratory.”
In 2018, a team from Children’s Hospital of Philadelphia, GrandOmics and three Chinese universities demonstrated that Bionano mapping on Saphyr can do all this. In the paper, they presented seven patients with FSHD1 whose genomes were mapped using Saphyr. In each case, the genome maps showed exactly how many copies of the repeats were present on each allele, identified the pathogenic allele and never let the chromosome 10 repeat array get in the way. On top of that, they demonstrated Bionano’s ability to resolve somatic mosaicism of the contracted repeat in two patients. All patient genomes were independently analyzed using two different labeling enzymes, and the results generated from both experiments correlated perfectly with each other, suggesting that a single Bionano run can provide you with the needed quality data.
In 2017, Professor Jonathan Pevsner from John Hopkins University and the Kennedy Krieger Institute presented his results on FSHD mapping with Bionano at the AGBT conference (Dr. Pevsner also studies autism genomics, more about that here). He presented equally successful results as shown in this paper, and he also reported that he tried to perform the same experiment using Illumina short-read sequencing and the barcoding approach from 10x Genomics. Neither method could resolve the D4Z4 region nor distinguish between the 4q and 10q loci. Short reads, whether they are linked or not, and repeat arrays just don’t go together well!
Later in 2017, Dr. Pevsner worked with Yuval Ebenstein’s team at Tel Aviv University to map the methylation status of the D4Z4 repeat in patients with FSHD as well.
All these studies make the same point: Bionano can see structural variants that other molecular methods can’t, and demonstrate yet again the power of ultra long molecules capable of spanning the most difficult regions of the genome.