Journal: Nature methods

Article Title: SVision: A deep learning approach to resolve complex structural variants

doi: 10.1038/s41592-022-01609-w

Figure Lengend Snippet: a, Overview of three modules in SVision. b, Workflow of each module; SOI- segment of interest. c, Performance for calling simulated complex structural variants (CSVs) was evaluated by recall (y-axis), precision (x-axis) and F-score (F, dashed line). d, SVision detected all previously reported and manually curated NA12878 CSVs, all of which passed exact match criteria. e, A misinterpreted complex event from short-read data, which is correctly detected by Svision as simple deletion. This event is surrounded by repeats that introduce two distinct patterns of discordant paired-end mapping, misleading short-read callers to report a CSV.

Article Snippet: Reconstructed: SVision CSV structure validated via contig based manual curation.

Techniques: Introduce, End-sequence Profiling

Journal: Nature methods

Article Title: SVision: A deep learning approach to resolve complex structural variants

doi: 10.1038/s41592-022-01609-w

Figure Lengend Snippet: a, One of the 12 inverted duplication events detected by SVision and classified as CSV graph structure ‘12’, b, One of the eight deletion associated with inversion events detected by SVison and classified as CSV graph structure ‘15’. c, One of the five multiple-deletion with spacer events detected by SVision and classified as CSV graph structure ‘27’. d, One of ten deletion with inverted duplication events detected by SVision and classified as CSV graph structure ‘23’. e, One of the five deletion with duplication events detected by SVision and classified as CSV graph structure ‘28’. From figure (a) to (e), the Dotplots on the left column are SVision one-variant images created with variant feature sequence (VAR, y-axis) and reference sequence (REF, x-axis) at the variant loci, while the Dotplots on right column are created with variant spanning HiFi assemblies (CONTIG, y-axis) and the reference sequence (REF, x-axis) at the variant loci.

Article Snippet: Reconstructed: SVision CSV structure validated via contig based manual curation.

Techniques: Variant Assay, Sequencing

Journal: Nature methods

Article Title: SVision: A deep learning approach to resolve complex structural variants

doi: 10.1038/s41592-022-01609-w

Figure Lengend Snippet: a, The Svision complex structural variants (CSVs) overlapping with Phased Assembly Variant (PAV) calls and reconstructed with phased HiFi contigs. The rare type represented multiple CSV graph structures, each of which contains fewer than five complex events. A: Inverted duplication. B: Deletion associated with inversion. C: Deletion associated with inverted duplication. D: Multiple deletion with spacer. E: Deletion associated with duplication. Inconclusive: Unable to characterize visually. Reconstructed: SVision CSV structure validated via contig based manual curation. Novel: No overlapping PAV call. b, Compared to 80 CSVs in HG00733 detected by SVision, short-read callset correctly reported the full structure of 4 CSVs (exact match), partially called 7 (region match), misinterpreted 23 events as simple (region match) and completely missed 46 CSVs. c, Three distinct alleles were found for the CNTN5 locus, including a contracted tandem repeat (REF), an expanded tandem repeat (INS), and an expanded tandem repeat with a complex duplication containing CNTN5 exon 4. d, The frequencies of three CNTN5 alleles differ among populations.

Article Snippet: Reconstructed: SVision CSV structure validated via contig based manual curation.

Techniques: Variant Assay

Journal: Nature methods

Article Title: SVision: A deep learning approach to resolve complex structural variants

doi: 10.1038/s41592-022-01609-w

Figure Lengend Snippet: a, An example of a complex structural variant (CSV) graph where its graph path is interpreted as S1+S3-S3-S4+. b, Examples of isomorphic graphs representing two different CSV events. c, SVision detected CSV classified as local target site duplication. d, SVision detected CSV classified as tandem duplication. Though events of structure depicted by (c) and (d) were computed as complex events, they were considered as simple events from the biological perspective.

Article Snippet: Reconstructed: SVision CSV structure validated via contig based manual curation.

Techniques: Variant Assay