GraphPed 0.0.5

A pedigree visualization tool based on dot format and visualized by graphviz

GraphPed

A novel graph-based visualization method for large and complex pedigrees

Author: Yin Huang

Citation:

Features

• It can deal with complex families, wrong pedigrees, and multiple groups in a family.

• It can help to check pedigrees.

• It can show multiple traits or status in one pedigree.

• It can show pedigrees in jupyter notebook and output as common image format (pdf, svg, png).

Install

pip install graphped

How to use

1. In command line

!GraphPed -h

usage: GraphPed [-h] [-p PED] [-o OUTPUT] [-f FORMAT] [-a ATTRIBUTES]
                [-e ENGINE]

The arguments of graphped

options:
  -h, --help            show this help message and exit
  -p PED, --ped PED     a ped file or an extended ped file (default: None)
  -o OUTPUT, --output OUTPUT
                        output folder (default: ./)
  -f FORMAT, --format FORMAT
                        the format of the output picture (default: svg)
  -a ATTRIBUTES, --attributes ATTRIBUTES
                        the attributes of the output picture (default: None)
  -e ENGINE, --engine ENGINE
                        the engine of graphviz rendering the output picture
                        (default: dot)

• 1. standard pedigrees in the ped file

GraphPed -p data/example_fam.ped -o data/cli/ -f pdf

• 2. extended pedigrees in the ped file

GraphPed -p data/example_fam_ext.ped -o data/cli/ -f svg -a data/default.yaml 

2.In jupyter notebook

from graphped.plot import *

fam=readped('data/example_fam.ped')
plotped(fam)

Or

show(GraphPed(fam))

GraphPed function can plot all the pedigrees in the fam dataframe.

Adding self-defined attributes. the number of traits in the input file should match with the number of traits in the attribute yaml file.

attrs=load_attributes('data/default.yaml')
famext=readped('data/example_fam_ext.ped',attrs)
plotped(famext,attrs)

Write to output folder with pdf format

plotped(famext,attrs,output='data/jpn',format='pdf')

Or output multiple pedigrees.

GraphPed(famext,attrs,output='data/jpn',format='pdf')

Tutorial

Setting the attribute yaml file

• reference:
  • fillcolor https://graphviz.org/docs/attrs/fillcolor/
  • style https://graphviz.org/docs/attrs/style/
  • fontcolor https://graphviz.org/docs/attrs/fontcolor/
  • ... https://graphviz.org/doc/info/attrs.html

For one trait ped file, if the trait values are affected status, which should be coded as follows: -9 or 0 is missing,1 is unaffected, and 2 is affected. you don't need to set the attribute file. Otherwise, you need to set your attribute file by following:

The format of the attribute of yaml file

trait name:
    attribute name:
        (the pairs of tait value and attribute value)
        tait value1: attribute value1
        tait value2: attribute value2
        ...

If you have more than one traits, you need to set each trait separately in the yaml file. The following is an example.

%%writefile data/default.yaml

trait1:
    fillcolor:
        1: 'white'
        2: 'dimgrey'
        -9: 'aquamarine3'

trait2:
    style:
        True: filled,setlinewidth(4)
        False: filled
    

trait3:
    fontcolor:
        True: darkorange
        False: black
    

Overwriting data/default.yaml

attrs=load_attributes('data/default.yaml')

attrs

{'trait1': {'fillcolor': {1: 'white', 2: 'dimgrey', -9: 'aquamarine3'}},
 'trait2': {'style': {True: 'filled,setlinewidth(4)', False: 'filled'}},
 'trait3': {'fontcolor': {True: 'darkorange', False: 'black'}}}

Two example pedigrees

one is standard, one is extended with 3 traits.

Standard ped file

%%writefile data/example_fam.ped
Fam	F4	P3	F1	1	1
Fam	F3	P3	F1	2	1
Fam	F2	P3	F1	2	1
Fam	F1	P1	P2	2	2
Fam	P3	0	0	1	2
Fam	P1	0	0	1	-9
Fam	P2	0	0	2	-9

Overwriting data/example_fam.ped

fam=readped('data/example_fam.ped')

fam

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	fid	iid	fathid	mothid	sex	trait
0	Fam	F4	P3	F1	1	1
1	Fam	F3	P3	F1	2	1
2	Fam	F2	P3	F1	2	1
3	Fam	F1	P1	P2	2	2
4	Fam	P3	0	0	1	2
5	Fam	P1	0	0	1	-9
6	Fam	P2	0	0	2	-9

plotped(fam)

plotped(fam)

Extended ped file

%%writefile data/example_fam_ext.ped
Fam1	F4	P3	F1	1	1	True	False
Fam1	F3	P3	F1	2	1	True	True
Fam1	F2	P3	F1	2	1	True	False
Fam1	F1	P1	P2	2	2	True	False
Fam1	P3	0	0	1	2	True	False
Fam1	P1	0	0	1	-9	False	True
Fam1	P2	0	0	2	-9	False	True

Overwriting data/example_fam_ext.ped

famext=readped('data/example_fam_ext.ped',attrs)

famext

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	fid	iid	fathid	mothid	sex	trait1	trait2	trait3
0	Fam1	F4	P3	F1	1	1	True	False
1	Fam1	F3	P3	F1	2	1	True	True
2	Fam1	F2	P3	F1	2	1	True	False
3	Fam1	F1	P1	P2	2	2	True	False
4	Fam1	P3	0	0	1	2	True	False
5	Fam1	P1	0	0	1	-9	False	True
6	Fam1	P2	0	0	2	-9	False	True

plotped(famext,attrs)

dots=GraphPed(fam)

show(dots)

Write out plots

plotped(fam,output='data/exampleplots',format='png')

Show the plot from data/exampleplots/Fam.png

Real data examples

all_fam=readped('data/Fig_2_3_fam.ped')

all_fam

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	fid	iid	fathid	mothid	sex	trait
0	25_2	25_2_33	25_2_49	25_2_50	2	2
1	25_2	25_2_28c1	0	0	2	1
2	25_2	25_2_25	25_2_49	25_2_50	1	2
3	25_2	25_2_28	0	0	1	2
4	25_2	25_2_29	25_2_49	25_2_50	2	2
...	...	...	...	...	...	...
401	10R_R99	10R_R99_22	10R_R99_29	10R_R99_15	1	1
402	10R_R99	10R_R99_29	0	0	1	2
403	10R_R99	10R_R99_7	0	0	2	2
404	10R_R99	10R_R99_8	0	0	1	1
405	10R_R99	10R_R99_8	10R_R99_19	10R_R99_5	2	1

406 rows × 6 columns

Fig.1 The workflow of GraphPed

show workflow

Fig.2 The pedigrees of complex families

plotped(all_fam[all_fam.fid=='25_2'])

Fig.S2 The largest pedigree in ADSP

plotped(all_fam[all_fam.fid=='4_649'])

Fig.3 The pedigrees with incorrect information

plotped(all_fam[all_fam.fid=='10R_R99'])

Fig.4 The pedigrees with multiple phenotypes

self-defined multiple-trait yaml

%%writefile data/self_defined_mutiple_traits.yaml

ad:
    fillcolor:
        1: 'white'
        2: 'dimgrey'
        -9: 'aquamarine3'

vcf:
    style:
        True: filled,setlinewidth(4)
        False: filled
    

trim:
    fontcolor:
        True: darkorange
        False: black
    

Writing data/self_defined_mutiple_traits.yaml

attrs=load_attributes('data/self_defined_mutiple_traits.yaml')
ped=readped('data/Fig4_fam_ext.ped',attrs)

ped

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	fid	iid	fathid	mothid	sex	ad	vcf	trim
0	10R_R99	10R_R99_10	10R_R99_2	10R_R99_1	2	1	True	False
1	10R_R99	10R_R99_19	10R_R99_2	10R_R99_1	1	1	True	False
2	10R_R99	10R_R99_20	10R_R99_2	10R_R99_1	1	1	False	False
3	10R_R99	10R_R99_21	10R_R99_2	10R_R99_1	1	1	False	False
4	10R_R99	10R_R99_17	10R_R99_29	10R_R99_15	1	1	True	False
5	10R_R99	10R_R99_22	10R_R99_29	10R_R99_15	1	1	False	False
6	10R_R99	10R_R99_15	10R_R99_8	10R_R99_7	2	2	True	False
7	10R_R99	10R_R99_12	10R_R99_8	10R_R99_7	2	2	True	False
8	10R_R99	10R_R99_1	10R_R99_8	10R_R99_7	2	2	True	False
9	10R_R99	10R_R99_5	10R_R99_8	10R_R99_7	1	1	True	False
10	10R_R99	10R_R99_6	10R_R99_8	10R_R99_7	2	1	True	False
11	10R_R99	10R_R99_2	0	0	1	-9	False	False
12	10R_R99	10R_R99_29	0	0	1	2	False	False
13	10R_R99	10R_R99_7	0	0	2	2	False	False
14	10R_R99	10R_R99_8	0	0	1	1	False	False

plotped(ped,attrs)

Show multiple figures with self-defined attributes

self-defined single-trait yaml

%%writefile data/self_defined_single_trait.yaml

ad:
    fillcolor:
        1: 'white'
        2: 'dimgrey'
        -9: 'aquamarine3'

Writing data/self_defined_single_trait.yaml

attr=load_attributes('data/self_defined_single_trait.yaml')
all_fam=readped('data/Fig_2_3_fam.ped',attr)

all_fam

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	fid	iid	fathid	mothid	sex	ad
0	25_2	25_2_33	25_2_49	25_2_50	2	2
1	25_2	25_2_28c1	0	0	2	1
2	25_2	25_2_25	25_2_49	25_2_50	1	2
3	25_2	25_2_28	0	0	1	2
4	25_2	25_2_29	25_2_49	25_2_50	2	2
...	...	...	...	...	...	...
401	10R_R99	10R_R99_22	10R_R99_29	10R_R99_15	1	1
402	10R_R99	10R_R99_29	0	0	1	2
403	10R_R99	10R_R99_7	0	0	2	2
404	10R_R99	10R_R99_8	0	0	1	1
405	10R_R99	10R_R99_8	10R_R99_19	10R_R99_5	2	1

406 rows × 6 columns

dots=GraphPed(all_fam,attr)
show(dots)