In the case of ITS1 sequences, there are differences among species: in N. tomentosiformis there is a single abundant ITS1 cluster (ribotype); however, in N. sylvestris, N. kawakamii and N. otophora there are multiple read-rich ITS1 clusters (2 3 of which exceed 5% of the total number of reads).
For each selected exon, we made sure that there were multiple reads spanning the selected exon with at least one other known exon.
For each position on each reference contig, for each individual, if there were multiple reads whose alignments started at the position, then only a single one selected uniformly at random was retained; this capped coverage to a high but computationally manageable level for GATK, while maintaining high diversity of contributing reads.
It is easy to see that 0 < c i ≤ 1 and if there are multiple supportive reads for an SV it is the maximum centeredness that matters the most (because the evidence best supportive of presence is the most informative).
In the genome sequencing data, there are multiple clear peaks of reads, suggesting the presence of collapsed repeats in the reference genome [the reference genome used throughout is hg18; ~10% of these regions are no longer problematic in hg19 (Supplementary Fig. S1)].
Additionally, if there are multiple variants under a read's mapping, the read may be mapped to multiple positions in the genome, but usually only the best mappings are reported.
The 'LCA Match' category means the LCA has a lower read label score but there are multiple conflicting labels with higher comparable read label score.
There are multiple important factors during the read library preparation phase of any NGS experiment that can affect the performance of a structural variation identification method.
The Dp140 promoter lies between exons 44 and 45 of the human dystrophin gene, and although the corresponding first exon splices onto exon 45, the first in-frame initiation codon is not until exon 51 (though there are multiple AUGs in other reading frames).
