A logical problem with SplitCommonSuffices and MergeCommonSuffices

ZexuanZexuan JiangsuChinaMember

@Sheila @valentin @depristo
For example :
A+x -> y (A+x,y is a point)
B+x -> y
after SplitCommonSuffices
A -> x -> y (A,B,x,y is a point)
B -> x -> y
after MergeCommonSuffices
A -> x => A+x
B -> x => B+x
then after SplitCommonSuffices after MergeCommonSuffices ...
of course some time after SplitCommonSuffices SeqVertex id is bigger than before but this may occured in unintended(unmind) circumstances.

Best Answer

Answers

  • SheilaSheila Broad InstituteMember, Broadie, Moderator

    @Zexuan
    Hi,

    I asked Valentin to comment. He should get back to you soon.

    -Sheila

  • vruanovruano Member
    edited January 22

    @Zexuan
    Sorry, I'm not sure I understand what is the question, doubt or comment.

    For example perhaps you could clearly state what you see after the operation (or operations) and what you actually see.

    Also when you try to represent graphs with with letter and arrows it important that you indicate whether two vertices with the same letter are actually the same vertex or are just two vertices that happen to contain the same base sequence; that makes a difference.

    Are the x (and y) that accessible from A and B the same vertex:

    A --> x --> y
    B --> x --> y
    

    or

    A --> x --> y
    B _/
    

    or

    A --> x --> y <-- x <-- B        
    
  • ZexuanZexuan JiangsuChinaMember
    edited January 23
  • vruanovruano Member
    edited January 23

    @Zexuan, Please accept my apologies if this may cause some frustration considering the PR containing the fix for the alleged bug is that simple, but you have to understand that we cannot merge in that PR until the dev team has a clear understanding of what the issue is.

    So please let me insist that you state what the actual problem is:

    1. Input graph.
    2. Operation sequence that reproduce the problem.
    3. Expected output graph.
    4. Actual output graph.
    5. Finally, if expected and actual are different, why that would be an issue (eg. is affecting variant discovered or calls or is just a performance issue [redundant nodes])

    Please make sure that in your graph text representation we can figure out whether two vertices with the same name (sequence) are the same object or two separate objects.

    You don't need to do this here, perhaps the PR that you open would be a more appropriate location for that since this is quite low level.

  • ZexuanZexuan JiangsuChinaMember
    Accepted Answer

    I just want this program run better...

Sign In or Register to comment.