4 ms·
And it's specifically the low error rate of the HiFi sequencing that enabled this advance. The string graph collapses not just because of sequence length, but e
by inciampati 5y ago
And it's specifically the low error rate of the HiFi sequencing that enabled this advance. The string graph collapses not just because of sequence length, but error rate that makes it impossible to distinguish repeats from each other. The 15kb reads used in this paper have just enough length and accuracy to span almost all repeats in the genome (rDNA and some centromeres excluded).
The assembly used the same conceptual foundation as the original human genome assemblies, for the curious: https://doi.org/10.1093/bioinformatics/bti1114 https://doi.org/10.1093/bioinformatics/bti1114
- DrAwdeOccarim 5y agoCan you say more about why rDNA is excluded? Does this stand for ribosomal DNA? Like the DNA that codes for rRNAs? Why is that difficult to sequence?
- callesgg 5y agoTwo guesses. rDNA folds itself so much that it becomes hard to read. Using the ribosomes to read ribosome dna might cause the ribosomes to bind up.
- tgb 5y agorDNA is repeated many many times and is located at several places in the genome. Presumably we need so many ribosomes that we need to have many copies of it or else transcribing it into rRNA would be the bottle neck in translating all mRNA. Can't have any proteins without rRNA!
- DrAwdeOccarim 5y agoCool, I didn't know that! Do you know how many times? Edit: 200 (http://book.bionumbers.org/how-many-ribosomal-rna-gene-copies-are-in-the-genome/ http://book.bionumbers.org/how-many-ribosomal-rna-gene-copie...) and its why nucleoli are a thing. This answers so many of my questions about the nucleoli and why so much RNA modification machinery is always found there...