VG format inputs and outputs about hisat2 HOT 3 OPEN

I'm not sure what "the paper" and "the authors" are. I also already know some other graph-based alignment programs. HISAT2 owes a great debt to GCSA. BWBBLE is probably not purely graph-based in my understanding - it uses 4 bits instead of 2 bits to represent all subsets of {A,C,G,T}.

There are a lot of algorithmic challenges for graph based alignment, so collaboration is really important. I'm definitely open to using some well organized graph formats like yours (vg), and your other suggestions sound good and promising to me. Please let me know how we proceed.

from hisat2.

Oh no, I'm sorry about this. I think I may have confused the HISAT preprint
with another article... Please disregard this.

In any case the salient point is about interoperability. To be of help, I
should take the system through its paces. If you'd like to think about
integration, the basic formats in vg are defined here:
https://github.com/ekg/vg/blob/master/vg.proto. Let me know if you have any
suggestions or elements that need to be added or modified to simplify
integration from your end. Anything that isn't clear should be specified in
the comments in the proto schema.
On Oct 21, 2015 10:07 AM, "Daehwan Kim" [email protected] wrote:

I'm not sure what "the paper" and "the authors" are. I also already know
some other graph-based alignment programs. HISAT2 owes a great debt to
GCSA. BWBBLE is probably not purely graph-based in my understanding - it
uses 4 bits instead of 2 bits to represent all subsets of {A,C,G,T}.

There are a lot of algorithmic challenges for graph based alignment, so
collaboration is really important. I'm definitely open to using some well
organized graph formats like yours (vg), and your other suggestions sound
good and promising to me. Please let me know how we proceed.

—
Reply to this email directly or view it on GitHub
https://github.com/infphilo/hisat2/issues/10#issuecomment-149907099.

from hisat2.

That's okay! - I'll try to take a close look at it next month, and perhaps we can have a chat later.

from hisat2.