!Gfa section

Used for analyzing a GFA file. This GFA file needs to be blunt, i.e. contain no overlaps. P-, W-lines and sqz-ified files are all supported. GFA files can be compressed using gzip.

Required parameters:

• graph: GFA file path
• count_type: what feature should be counted inside the graph, one of Node, Bp, Edge (previously All was supported, but that has been removed as of panacus v0.5.0!)
• analyses: what analyses should be run, for this see their corresponding wiki pages.

Optional parameters:

• subset: what parts of the graph should be included, default is all, if given this includes every feature that is touched at least once by the annotations in the subset bed file
• exclude: what parts of the graph should be excluded, default is none, if given this excludes every feature that is touched at least once by the annotations in the subset bed file
• grouping: how paths should be grouped together, one of Sample, Haplotype, Custom. Sample and Haplotype use the PanSN-spec for naming when the file contains P-lines. Custom takes a two column .tsv-file with one column containing paths and the other the corresponding groups.
• reference: this contains the reference path that is used for assigning positions to features. This is only used in the regional analyses (!RegionalVariation, !RegionalGrowth, ...). Should be in PanSN-spec format and of the same level as grouping, e.g. CHM13#0 if grouping: Haplotype is set.
• nice: this can be set if segment IDs in the GFA file are strictly numerically increasing, without skipping any numbers, starting from 1 and going up to N, where N is the number of segments. Setting this speeds up GFA parsing massively, as there is no lookup for an internal ID when parsing paths/walks. However, please note that this is not true for most GFA graphs!

- !Gfa
  graph: ../graphs/my_graph_1.gfa
  grouping: Haplotype
  subset: ../graphs/my_graph_1.genes.bed
  exclude: ../graphs/my_exclude.bed
  reference: CHM13#0
  nice: false
  count_type: Bp
  analyses: