Closed SAMtoBAM closed 1 year ago
Hi @SAMtoBAM,
I don't think such a tool exists, yet. Amongst other things odgi depth
offers
-d, --graph-depth-table Compute the depth and unique depth on
each node in the graph, writing a
table by node: node.id, depth,
depth.uniq.
-v, --graph-depth-vec Compute the depth on each node in the
graph, writing a vector by base in one
line.
-D, --path-depth Compute a vector of depth on each base
of each path. Each line consists of a
path name and subsequently the
space-separated depth of each base.
-a, --self-depth Compute the depth of the path versus
itself on each base in each path. Each
line consists of a path name and
subsequently the space-separated depth
of each base.
-S, --summarize Provide a summary of the depth
distribution in the graph, in a
tab-delimited format it prints to
stdout: node.count, graph.length,
step.count, path.length,
mean.node.depth
(step.count/node.count), and
mean.graph.depth
(path.length/graph.length).
Would that suffice for you?
With number of paths you mean the unique number of paths crossing the node or all steps of all paths visiting the node?
Thanks for the response
Nicely we both converged on odgi depth
and it works rather well (using the -d option)
and I was looking for the unique number of paths so that provides the required information
I also converted the graph to a gfa (using odgi view
) and used that to extract the node length
I also used the gfa to confirm the results of odgi depth
so I have some peace of mind!
Also also used odgi pav
to extract similar information (using small windows along each path rather than nodes as the nodes could be problematic), giving me essentially the path coverage per window (I used it in terms of binary presence/absence). I got all that direction from here which was extremely helpful
Although I guess it is path-based instead of node-based, which may turn out to be easier for me in the long run.
Thanks again
Glad to hear that our tools and Docs were able to help you out!
Hi there,
Perhaps there is a simple way I have not found but I would like to essentially extract a list of all the nodes, there graph relative position, their length and coverage (number of paths). Is this possible?
I am looking for regions that are not present within all the haplotypes used to build the graph.
Thanks