Скачать или смотреть 20200504 Bioinformatics Sequencing Mapping Assembly

Slides for this lecture can be downloaded here: https://drive.google.com/file/d/11Brq...

My initial lecture for the bioinformatics of DNA sequencing discusses some of the most widely used bioinformatics strategies with sequencing data: basecalling error probabilities via Phred, mapping reads via the Burrows-Wheeler Aligner (BWA), and assembling contigs and constructing scaffolds via de Bruijn graphs built from k-mers. We spent a fair amount of time comparing technologies for Sanger sequencing and for massively parallel sequencing (more frequently called "Next Generation Sequencing"). In discussing the Phred scoring approach, we mentioned the value of Fourier Transform for eliciting electropherogram peak frequencies and the widely used negative log expression of low probability events (in this case, the probability of error associated with a basecall). Our relatively brief discussion of sequencing mapping discussed the value of suffix arrays for doing a single-pass assessment of all read positions in the annotated sequence. In our discussion of de novo assembly, we talked about the k-mer approach to discovering overlaps among different reads as well as the "Eulerian graph" approach to traversing along edges from one k-mer to another. We also mentioned the challenge posed by repetitive DNA sequences (frequently resulting from repeated sequences like Alu elements); these repeats can fool an assembler into wrongly juxtaposing two distant genes if they both flank similar repeats.