Fig. 2: The main steps of iGDA.
From: Detecting and phasing minor single-nucleotide variants from long-read sequencing data
Details are in the “Methods” section. a Encoding reads by using a single integer to represent both locus and identity of each substitution. b Generating subspaces by pairwise comparison of reads. c An illustrative example of the Random Subspace Maximization algorithm (RSM). d Detecting orphan SNVs by correcting sequence-context effect learnt from independent data. e Realigning each read to reduce reference bias. f An illustrative example of the Adaptive-Nearest Neighbor clustering algorithm (ANN). g Filtering contigs by frequencies and correlations of SNVs and similarities between contigs. h Assembling filtered contigs by overlap graph.
