Sub-package Documentation
Types of Output Files
expected name/suffix | file type/format | content |
---|---|---|
annotations.tab |
text/tabbed | annotated events |
annotations.fusion-cdna.fa |
fasta | putative fusion unspliced cDNA sequences |
drawings/*.svg |
SVG | diagrams |
drawings/*.legend.json |
JSON | diagram legend/metadata |
Algorithm Overview
- read in breakpoint pairs
- generate strand-specific annotations (one annotation per strand, multiple if multiple genes/transcripts in the region)
- try building fusion transcripts for bp-specific calls
- generate SVG diagrams
Levels of Annotations
Overview of Class Relationships
The Annotation sub-package has objects for genetic annotations and related calculations. The basic layout of the package is shown above. IS-A relationships are given by the blue arrows. HAS-A relationships are shown in black. And reference_object/parent type relationships are shown in red. mavis.annotate.genomic.Gene is a gene. Start and end are genomic positions wrt to the template/chr. mavis.annotate.genomic.PreTranscript is the unspliced transcript. Start and end are genomic positions wrt to the template/chr. mavis.annotate.genomic.Transcript: is the spliced transcript. Start and end coordinates are 1 to the length of the spliced product in base pairs. mavis.annotate.protein.Translation: is the translation of the spliced transcript. Start and end are cdna positions wrt the 5' end of the spliced transcript. The start and end here describe the start and end of the coding sequence