二叉树比较从叶到根python ete2工具的节点特征

Question

I have a binary tree:

       root
     /     \
    g      h
   / \    / \
 d   a   e   f
 / \
b   c

each node has 'seq' feature which stores the dna sequence of each node ('AACCGGT')

the sisters who are both leaves (b,c and e,f) have each one a score (float ) what I want is to :

• compare score of leaf which has score with it's sister b.score with c.score

• d.score = max (b.score,c.score)

• same for e, f and h

• compare a.seq with d.seq ==> d and e will have a score
• g.score max (e.score, a.score) ... till arrive to the root NOte: every leaf node has 'model feature' I compare b.seq with c.seq based on b.model ==>I got b.score then based on c.model ==> I got c.score

This the function I wrote but I'm not sure if it does what I want and I can't test it because I don't have the align_trna function yet

def affect_score(n):

if (n.score)==0:
            n.score,n.model=affect_score(n.get_children()[0])
        result=n.score
        model=n.model
        if not n.is_root():
            sis=n.get_sisters()[0]
            if sis.score==0:
                sis.score,sis.model=affect_score(sis.get_children()[0])
                n.score=align_trna(n.seq,sis.seq,n.model)
                sis.score,sis.model= align_trna(nseq, sis.seq,sis.model)
                if n.score < sis.score:
                        result=sis.score
                        model=sis.model

        return result,mode

l

Can anyone helps me by telling if I am thinking write ? Note that It's my first time working with tree data sturcture and recursion thanks in advance for any suggestion

Answer 1

Ok let's try to recap something about recursion.

The first thing to write when doing recursive function is the "exit condition" (the things that terminates the recursion at some point). In your case the exit condition should refer to the leaf node of the tree, and it would be something like:

if len(n.get_childreen())==0:
    return n.score

then for every other node you should do your computation:

child_score = max(affect_score(n).get_children()[0], affect_score(n).get_children()[1])
seq = compare_seq(n.seq, n.get_sisters()[0].seq)
n.score = max(child_score, seq)

then you also want a special condition on the root since it doesn't have any sisters:

if n.is_root():
   n.score = max(affect_score(n).get_children()[0], affect_score(n).get_children()[1])

At the end it should look like:

def affect_score(n):
    if len(n.get_childreen())==0:
        return n.score
    if n.is_root():
       n.score = max(affect_score(n).get_children()[0], affect_score(n).get_children()[1])
    else:
       child_score = max(affect_score(n).get_children()[0], affect_score(n).get_children()[1])
       seq = compare_seq(n.seq, n.get_sisters()[0].seq)
       n.score = max(child_score, seq)

This is what I understood you want to do! You need to add the compare_seq() which is the method that should compare the two sequences, and probably adapt something to your code. But in theory it should be almost correct.