gneiss.sort.gradient_sort

gneiss.sort.gradient_sort(tree, gradient, ascending=True)

Sorts tree according to ordering in gradient.

Parameters:

• tree (skbio.TreeNode) – Input tree where leafs correspond to features contained in the index in gradient.
• gradient (pd.Series, numeric) – Gradient where the index correspond to feature names. The index in the gradient must be consistent with names of the tips in the tree.

Returns:

A tree whose tips are sorted along the gradient.

Return type:

skbio.TreeNode

Examples

>>> from skbio import TreeNode
>>> import pandas as pd
>>> tree = TreeNode.read([u'((a,b)c, ((g,h)e,f)d)r;'])
>>> x = pd.Series({'f':3, 'g':1, 'h':2, 'a':4, 'b':5})
>>> print(tree.ascii_art())
                    /-a
          /c-------|
         |          \-b
-r-------|
         |                    /-g
         |          /e-------|
          \d-------|          \-h
                   |
                    \-f
>>> res = gradient_sort(tree, x)
>>> print(res.ascii_art())
                              /-g
                    /e-------|
          /d-------|          \-h
         |         |
-r-------|          \-f
         |
         |          /-a
          \c-------|
                    \-b