-
Notifications
You must be signed in to change notification settings - Fork 8
Wish List
Lydia Buntrock edited this page Nov 3, 2020
·
1 revision
Feature | SVIM | SVIRL | Vaquita | SVIM + SVIRL wishes |
---|---|---|---|---|
Considers signatures in CIGAR | yes | yes | yes | yes |
Considers signatures between alignments | yes | yes | yes | yes |
Variant types | Del, Ins, Inv, Dup (like Trans), TDup (Overlapping Reads) | currently only Del, breakends planned | Deletion, Inversion, Tandem Duplication, Duplication, Translocation (Inter-,intra chromosome), Breakends | Ins, breakends (= novel junctions); in later, separate step define also Del, Inv, Dup (wie Trans), TDup (Overlapping Reads) |
Distance metric of two reads | Length and position | Alignment of patches | Length and position | Both - user configurable |
Clustering approach | Hierarchical clustering of length and position | Vertex cover in conflict graph | Clustering based on positions | Both - user configurable |
Consensus computation | no, considered for insertions | yes | Yes but optional (not that effective I guess..) https://academic.oup.com/bioinformatics/article/25/9/1118/204548 | yes, but optional |
Output sequence in VCF | no | no | yes | yes |
Genotyping | yes | yes | Yes but weak (just based on depth, I have to check this..) | yes |
Language | Python with PySam | C++ with SeqAn2 | C++, SeqAn2 | C++ with SeqAn3 |
Step | Task | Output |
---|---|---|
1 | Identify both types of signatures as candidate breakends in reads (one candidate breakend corresponds to one read, a read can produce several candidate breakends) | Candidate breakends |
2 | Sliding window (output: subsets of candidate breakpoints) | Candidate breakends with subset number |
3 | Distance computation between pairs of candidate breakends (two approaches) per subset | Candidate breakends with subset number and pairwise distances in subset |
4 | Clustering (two approaches) | Candidate breakends with subset number and cluster number |
5 | (Optional): Consensus computation | Candidate breakends with subset number, cluster number and consensus |
6 | (Optional): Split alignment of consensus to reference | Candidate breakends with subset number, cluster number, consensus and alignment |
7 | Determining breakend position | Breakends in VCF format |
8 | Genotyping (two approaches) | Breakends in VCF format with genotypes |