browser

Contains the GenomeBrowser and GenomeStack classes

source

GenomeBrowser

 GenomeBrowser (gff_path:str=None, fasta_path:str=None, gb_path:str=None,
                seq_id:str=None, init_pos:int=None, init_win:int=10000,
                bounds:tuple=None, max_interval:int=100000,
                show_seq:bool=True, search:bool=True,
                attributes:Union[list,Dict[str,Optional[list]]]=None,
                feature_name:Union[str,Dict[str,str],NoneType]=None,
                feature_types:list=None, glyphs:dict=None, height:int=150,
                width:int=600, label_angle:int=45,
                label_font_size:str='10pt', label_justify:str='center',
                label_vertical_offset:float=0.03,
                label_horizontal_offset:float=-5, show_labels:bool=True,
                feature_height:float=0.15,
                features:pandas.core.frame.DataFrame=None,
                seq:Bio.Seq.Seq=None, color_attribute:str=None,
                z_stack:bool=False, **kwargs)

Initialize a GenomeBrowser object.

Type Default Details
gff_path str None path to the gff3 file of the annotations (also accepts gzip files)
fasta_path str None path to the fasta file of the genome sequence
gb_path str None path to a genbank file
seq_id str None id of the sequence to load, for genomes with multiple contigs, defaults to the first sequence in the genbank or gff file.
init_pos int None initial position to display
init_win int 10000 initial window size (max=20000)
bounds tuple None bounds can be specified. This helps preserve memory by not loading the whole genome if not needed.
max_interval int 100000 maximum size of the field of view in bp
show_seq bool True creates a html div that shows the sequence when zooming in
search bool True enables a search bar
attributes Union None list of attribute names from the GFF attributes column to be extracted. If dict then keys are feature types and values are lists of attributes. If None, then all attributes will be used.
feature_name Union None attribute to be displayed as the feature name. If str then use the same field for every feature type. If dict then keys are feature types and values are feature name attribute.
feature_types list None list of feature types to display
glyphs dict None dictionary defining the type and color of glyphs to display for each feature type
height int 150 height of the annotation track
width int 600 width of the inner frame of the browser
label_angle int 45 angle of the feature names displayed on top of the features
label_font_size str 10pt font size fo the feature names
label_justify str center center, left
label_vertical_offset float 0.03 how far above a feature to draw the label
label_horizontal_offset float -5 how far to shift the feature label on the x-axis
show_labels bool True if False, then don’t show feature labels
feature_height float 0.15 fraction of the annotation track height occupied by the features
features DataFrame None DataFrame with columns: [“seq_id”, “source”, “type”, “start”, “end”, “score”, “strand”, “phase”, “attributes”], where “attributes” is a dict of attributes.
seq Seq None keeps the Biopython sequence object
color_attribute str None feature attribute to be used as patch color
z_stack bool False if true features that overlap will be stacked on top of each other
kwargs

Additional keyword arguments are passed as is to bokeh.plotting.figure

Upon initialization a GenomBrowser object parses the data and creates a the GenomeBrowser.patches pandas DataFrame that contains the data to be plotted.

from genomenotebook.data import get_example_data_dir
import os
data_path = get_example_data_dir()
fasta_path = os.path.join(data_path, "MG1655_U00096.fasta")
gff_path = os.path.join(data_path, "MG1655_U00096.gff3")

g=GenomeBrowser(gff_path=gff_path, fasta_path=fasta_path, bounds=(0,50000))
print(g.seq_id, g.seq[:10])
g.patches.head()
U00096.3 AGCTTTTCAT
names xs ys xbox_min color alpha pos attributes type label_y label_x
0 thrL (190, 190, 190, 255, 190) (0.05, 0.2, 0.2, 0.125, 0.05) 190 purple 0.8 222.5 <span style="color:FireBrick">CDS</span><br><s... CDS 0.23 222.5
1 thrA (337, 337, 2699, 2799, 2699) (0.05, 0.2, 0.2, 0.125, 0.05) 337 purple 0.8 1568.0 <span style="color:FireBrick">CDS</span><br><s... CDS 0.23 1568.0
2 thrB (2801, 2801, 3633, 3733, 3633) (0.05, 0.2, 0.2, 0.125, 0.05) 2801 purple 0.8 3267.0 <span style="color:FireBrick">CDS</span><br><s... CDS 0.23 3267.0
3 thrC (3734, 3734, 4920, 5020, 4920) (0.05, 0.2, 0.2, 0.125, 0.05) 3734 purple 0.8 4377.0 <span style="color:FireBrick">CDS</span><br><s... CDS 0.23 4377.0
4 yaaX (5234, 5234, 5430, 5530, 5430) (0.05, 0.2, 0.2, 0.125, 0.05) 5234 purple 0.8 5382.0 <span style="color:FireBrick">CDS</span><br><s... CDS 0.23 5382.0

source

GenomeBrowser.show

 GenomeBrowser.show ()

Shows the plot in an interactive Jupyter notebook

#GFF + FASTA input
g=GenomeBrowser(fasta_path=fasta_path, gff_path=gff_path, bounds=(0,50000),width=600)
g.show()
#GenBank input
gb_path=os.path.join(data_path, "colored_genbank.gb")
g=GenomeBrowser(gb_path=gb_path)
g.show()
#Providing GFF file as the only input
g=GenomeBrowser(gff_path)
g.show()

source

GenomeBrowser.add_track

 GenomeBrowser.add_track (height:int=200, tools:str='xwheel_zoom,
                          ywheel_zoom, pan, box_zoom, save, reset',
                          **kwargs)

Adds a track to the GenomeBrowser. Ensures that the x_range are shared and figure widths are identical.

Type Default Details
height int 200 size of the track
tools str xwheel_zoom, ywheel_zoom, pan, box_zoom, save, reset comma separated list of Bokeh tools that can be used to navigate the plot
kwargs
Returns Track 
fasta_path = os.path.join(data_path, "MG1655_U00096.fasta")
gff_path = os.path.join(data_path, "MG1655_U00096.gff3")

data=pd.DataFrame(dict(x=np.arange(0,50000,100),
                       y=np.sin(np.arange(0,50000,100))))

g=GenomeBrowser(fasta_path=fasta_path, gff_path=gff_path, bounds=(0,5000), search=False, show_seq=False)

track = g.add_track(height=100)
track.scatter(data=data,pos="x",y="y")
g.show()

source

GenomeBrowser.highlight

 GenomeBrowser.highlight (data:pandas.core.frame.DataFrame=None,
                          left_col:str='left', right_col:str='right',
                          color_col:str='color', alpha_col:str='alpha',
                          left=None, right=None, color='green',
                          alpha:str=0.2, hover_data:List=None, **kwargs)
Type Default Details
data DataFrame None pandas DataFrame containing the data
left_col str left name of the column containing the start positions of the regions
right_col str right name of the column containing the end positions of the regions
color_col str color name of the column containing color of the regions
alpha_col str alpha name of the column containing alpha (transparency) of the regions
left NoneType None
right NoneType None
color str green
alpha str 0.2 transparency
hover_data List None list of additional column names to be shown when hovering over the data
kwargs 
import pandas as pd
highlight_regions=pd.DataFrame({"start": [5000, 8000], "stop": [6000, 8500], "color": ["red","green"], "y":[23, 45]})

g=GenomeBrowser(gff_path=gff_path, fasta_path=fasta_path, bounds=(0,10000))
g.highlight(data=highlight_regions, left_col="start", right_col="stop", hover_data=["y"])
g.show()
data=pd.DataFrame(dict(x=np.arange(0,50000,100),
                       y=np.sin(np.arange(0,50000,100))))

g=GenomeBrowser(fasta_path=fasta_path, gff_path=gff_path, bounds=(0,5000), search=False, show_seq=False)
track = g.add_track(height=100)
track.scatter(data=data,pos="x",y="y")

highlight_regions=pd.DataFrame({"start": [2000, 4000], "stop": [3000, 4500], "color": ["red","green"], "y":[23, 45]})
g.highlight(data=highlight_regions, left_col="start", right_col="stop", hover_data=["y"], highlight_tracks=True)

g.show()

source

GenomeBrowser.add_tooltip_data

 GenomeBrowser.add_tooltip_data (name:str, values:str,
                                 feature_type:str=None)
Type Default Details
name str name of the data to be added
values str values
feature_type str None specify the feature type if the data applies only a to specific feature_type 
g=GenomeBrowser(gff_path=gff_path, attributes=["gene", "locus_tag"],bounds=(0,10000))
data=np.random.randint(0,10,len(g.patches))
g.add_tooltip_data("data",data)
g.show()

source

GenomeBrowser.save

 GenomeBrowser.save (fname:str, title:str='Genome Plot')

Saves the plot in svg or png. This function saves the initial plot that is generated and not the current view of the browser. To save in svg format you must initialise your GenomeBrowser using output_backend="svg"

Type Default Details
fname str file name (must end in .svg or . png).If using svg, GenomeBrowser needs to be initialized with output_backend="svg"
title str Genome Plot plot title

Saving to svg

g=GenomeBrowser(gff_path=gff_path, 
                bounds=(0,5000),
                search=False)
track = g.add_track(height=100)
track.scatter(data=pd.DataFrame(dict(x=np.arange(0,5000,100),y=np.sin(np.arange(0,5000,100)))), y="y", pos="x")
g.save("test.svg")
g.show()

Saving to png

g=GenomeBrowser(fasta_path=fasta_path, 
                gff_path=gff_path,
                bounds=(0,5000),
                search=False,
                height=200,
                width=2000,
                label_font_size="20pt")
g.save("test.png")

source

GenomeStack

 GenomeStack (browsers=None)

Initialize self. See help(type(self)) for accurate signature.

data_path = get_example_data_dir()
gb_path = os.path.join(data_path, "colored_genbank.gb")
g = GenomeStack.from_genbank(gb_path,
                                width=700, 
                                show_seq=True, 
                                search=False, 
                                feature_types=["CDS", "Domainator"], 
                                color_attribute="Color", 
                                label_angle=0, 
                                show_labels=True, 
                                feature_height=0.15, 
                                label_vertical_offset=-0.15, 
                                label_justify="left", 
                                label_horizontal_offset = 5, 
                                glyphs=get_default_glyphs(box_colors=("Cyan",)), 
                                feature_name={"CDS":"gene_id","Domainator":"name"},
                                z_stack=True) 
g.show()