If you're viewing this notebook on github and don't see the pie chart at the end, view it on the jupyter nbviewer instead.

This is a trivial example demonstrating how to connect to a ga4gh server, select annotations from one annotation set, and generate a pie chart from Sequence Ontology terms associated with all variants. When applied to gene coordinates and a specific transcript, a similar query would correspond to the observed mutation spectrum.

The original query was lifted from David Steinberg.



In [1]:

    
import collections

from plotly import __version__
from plotly.offline import download_plotlyjs, init_notebook_mode, iplot
init_notebook_mode()
import plotly.graph_objs as go

import ga4gh.client as client

args=dict(
    variantAnnotationSetId = "YnJjYTE6T1I0Rjp2YXJpYW50YW5ub3RhdGlvbnM",
    referenceName = "1",
    start = 0,
    end = 300000000
    )









    



INFO:requests.packages.urllib3.connectionpool:Starting new HTTPS connection (1): api.plot.ly



In [2]:

    
gc = client.HttpClient("http://localhost:8000")



In [3]:

    
annotation_set = gc.getVariantAnnotationSet(args["variantAnnotationSetId"])



In [4]:

    
c = collections.Counter(
    ef.term
    for va in gc.searchVariantAnnotations(**args)
    for te in va.transcriptEffects   
    for ef in te.effects)



In [5]:

    
labels, values = zip(*c.most_common())
fig = {
    'data': [{'labels': labels, 'values': values, 'type': 'pie'}],
    'layout': {'title': """Transcript effects for {a[referenceName]}:{a[start]}-{a[end]}"""
               """<br>variant annotation set {a[variantAnnotationSetId]}""".format(a=args)}
}
iplot(fig)



In [ ]: