The cause of rare diseases is increasingly being detected through genome sequencing, which involves reading the entire human DNA by first breaking it into small pieces—short reads. Christian Gilissen, Lisenka Vissers, and colleagues found that a new technique using long reads is even more effective at detecting complex causes. They report that eighty to ninety percent of cases were detectable, as stated in the American Journal of Human Genetics .
Rare diseases are typically due to genetic causes. These causes are more and more often identified using genome sequencing techniques. Genome sequencing examines the entire genome of an individual, which consists of over 3 billion building blocks of DNA.
Missed Information
To read a genome, DNA is first cut into small pieces of about 300 letters, known as short reads. These pieces are read and then reassembled until the complete genome is mapped. "This method has allowed us to identify the cause of quite a few rare diseases," says Christian Gilissen, Professor of Genome Bioinformatics at Radboudumc. However, the short reads technique also has its drawbacks. The huge collection of small genetic pieces cannot always be reassembled in the exact right places. This is especially true for very long repetitive sections of DNA (repeats) and for pieces of DNA that have been deleted, inserted, or rearranged (translocations). To compensate for these issues with short reads, several other techniques are often employed to detect what may have been missed.
Long Reads
In recent years, a new sequencing technique based on long reads has emerged, by PacBio HiFi sequencing among others. With this technique, genetic segments of 20,000 letters can be read, as opposed to the 300 letters of short reads. Vissers, Professor of Translational Genomics at Radboudumc explains, "Initially, this technique was less accurate and quite expensive, but it has now become reliable and much more affordable. Therefore, we asked ourselves: could we replace the short reads and all the additional tests with long reads? Would long reads be able to capture enough genetic mutations so that we could replace the short reads and other supplementary tests in the clinic with long reads?"
Over Ninety Percent
Gilissen and colleagues selected 100 samples where the genetic causes of rare diseases had been difficult to identify in previous investigations using short reads combined with various supplementary tests. These samples were now solely examined using long reads. Gilissen says, "With long reads, we immediately identified 83 percent of the causes. Upon further examination, we found that an additional 10 percent had been detected by the technique but were not automatically identified. For the remaining 7 percent, further improvements in the technique may be necessary. Essentially, we are now picking up 93 percent of the causes in this more challenging test. Causes that cannot, or are very difficult to detect with current short read techniques. For less difficult causes, this percentage is likely even higher."
Switching Genes On and Off
Long reads also have the advantage of directly mapping methylation. Methylation is a process through which genes are turned on or off via chemical changes to the DNA. Gilissen says, "Short reads cannot do that, and you need an extra test for it. We included an example of this in our samples. Normally, the mother's gene is active and the father's gene inactive, but in this case the patient received no gene from the father and two genes from the mother. Both of the mother's genes were inactive. This was the cause of this rare disease. It's a bit complicated, but we were able to immediately identify such a rare and complex example."
One Test
In recent years, more research has been conducted on the use of long reads for clinical applications. According to Vissers, this is the first time such a direct comparison has been made between 100 already known and difficult-to-find causes of rare diseases and the use of long reads. With positive results, she says. "With further improvements in the technique and falling prices, in the near future long reads seem to be the preferred method for rare disorders in terms of quality, ease of use —one test instead of several — and affordability."
----------------------------
Paper in The American Journal of Human Genetics: HiFi long-read genomes for difficult-to-detect clinically relevant variants - Wolfram Höps, Marjan M. Weiss, Ronny Derks, Jordi Corominas Galbany, Amber den Ouden, Simone van den Heuvel, Raoul Timmermans, Jos Smits, Tom Mokveld, Egor Dolzhenko, Xiao Chen, Arthur van den Wijngaard, Michael A. Eberle, Helger G. Yntema, Alexander Hoischen, Christian Gilissen, Lisenka E.L.M. Vissers
