your question is a little vague on what exactly you are trying to do and a code example would help give a better answer.
Generally when trying to plot geographical data obtained by online sources, you will have data in a coordinate system (WGS84) with latitude being the y and longitude being the x. Bokeh can plot longitude and latitude simply by specifying the proper names in the ColumnDataSource.
from bokeh.io import show
from bokeh.models import ColumnDataSource
from bokeh.plotting import figure
longitude = [44.990961]
latitude = [41.552164]
source = ColumnDataSource(data=dict(longitude=longitude, latitude=latitude))
p = figure(plot_width=400, plot_height=400)
p.circle(x='longitude', y='latitude', source=source)
show(p)
If truly your question is related to a coordinate transformation of your data, it will be difficult to answer your question without more details. I recommend you take a look at https://en.wikipedia.org/wiki/Map_projection to understand map projections.
If you need to convert the longitude/latitude coordinates into a different coordinates system. You can use the pyproj
package.
import pyproj
project_projection = pyproj.Proj("+init=EPSG:4326") # wgs84
google_projection = pyproj.Proj("+init=EPSG:3857") # default google projection
longitude = [44.990961]
latitude = [41.552164]
x, y = pyproj.transform(google_projection, project_projection, longitude, latitude)
print(x, y)
Reference for the google projection Google map api v3 projection?