PyAGH

Introduction

PyAGH is a Python package developed for calculating relationship matrix using pedigree, genotype or microbiology data as well as processing, analysis and visualization for data. PyAGH provides fast and concise methods for calculating Amatrix (based on pedigree), Gmatrix (based on genotype), Mmatrix(based on OTU) and Hmatrix (based on pedigree and genotype) used in breeding. PyAGH supports high marker density typing data, large pedigree data, microbiome data, additive, dominant and epistatic effects relationship matrix, relationship matrix construction across population. With the obtained relationship matrix, you can use PyAGH for fast visualization, including PCA analysis, Cluster analysis and drawing Heatmaps.

To make it easier to use pedigree information, PyAGH also provides targeted tools for specific needs, such as detecting pedigree errors, selection individuals, pedigree visualization, sorting pedigree, calculating inbreeding coefficients and calculating relationship coefficients . The latest stable release of the software can be installed conveniently via pip .

Target Audience

The target audience of PyAGH includes:

• Students and researchers in the field of breeding and genetics, particularly those who want to perform genomic prediction.
• Government, enterprises, or other entities who expect efficient processing of pedigree and genomic information.

Technical Features

• Provide a variety of methods for calculating kinship matrix including combined reference population.
• Fast and support tens of millions of genotypes.

Main Functions

Currently, PyAGH mainly provides the following methods:

• Pedigree sorting: Provides methods to quickly obtain the correct pedigree in chronological order of birth clean multiple types of data error.
• Pedigree selection: Provides methods to select pedigree based on specific individuals and generations.
• Calculation of coefficients: Provides methods to easily obtain the inbreeding and relationship coefficients of specific individuals.
• Kinship matrix: Provides different methods to calculating Amatix, Gmatrix and Hmatrix.
• Visualization: Provides methods to display results.

---

Installation

It is recommended to use Python 3.9

Using pypi

PyAGHcan be installed by using pip install.

pip install PyAGH

Examples

Load the example data

loadEgPed() and loadEgGeno() functions can load the example data.

import PyAGH
ped = PyAGH.loadEgPed()
genofile = PyAGH.loadEgGeno() #is a filename
ped

index	id	sire	dam
0	9	1	2
1	10	3	4
2	11	5	6
3	12	7	8
4	13	9	10
5	14	11	12
6	15	11	4
7	16	13	15
8	17	13	14

Sort pedigree

sortPed() is a function that can sort the pedigree according to the correct birth date of individuals and check for various errors in the pedigree.

ped_sorted = PyAGH.sortPed(ped)
ped_sorted

index	id	sire	dam
0	1	0	0
1	3	0	0
2	5	0	0
3	7	0	0
4	2	0	0
5	4	0	0
6	6	0	0
7	8	0	0
8	9	1	2
9	10	3	4
10	11	5	6
11	12	7	8
12	15	11	4
13	13	9	10
14	14	11	12
15	16	13	15
16	17	13	14

Select pedigree

selectPed() function can select pedigree based on specific individuals and generations.

ped_selected = PyAGH.selectPed(data=ped,id=['9','10'],generation=3)
ped_selected

index	id	sire	dam
0	1	0	0
1	3	0	0
2	2	0	0
3	4	0	0
4	9	1	2
5	10	3	4

Calculate kinship matrix

makeA(), makeG() and makeH() functions can calculate the kinship matrix using various methods based on pedigree , genotype and both together, respectively.

A = PyAGH.makeA(ped_sorted)
G = PyAGH.makeG(File=genofile,method=1,File_list=False)
G_inter = PyAGH.makeG_inter(geno,method="dd")
M = PyAGH.makeM(OTU)
H = PyAGH.makeH(G,A,w=0.05)
####A,G and H are lists with 2 elements. The first one is kinship matrix in numpy.ndarray type and the other one is id labels in pandas.Series type. 

Calculate coefficients

coef_inbreeding = PyAGH.coefInbreeding(A)
coef_inbreeding

index	ID	F
0	1	0.0000
1	3	0.0000
2	5	0.0000
3	7	0.0000
5	4	0.0000
11	12	-0.0000
...	...	...
12	15	-0.0000
13	13	-0.0000
14	14	-0.0000
15	16	0.0625
16	17	-0.0000

coef_kinship = PyAGH.coefKinship(A)
coef_kinship

	ID1	ID2	r
0	1	1	1.000000
1	1	3	0.000000
2	1	5	0.000000
3	1	7	0.000000
4	1	2	0.000000
...	...	...	...
148	14	16	0.121268
149	14	17	0.500000
150	16	16	1.000000
151	16	17	0.333486
152	17	17	1.000000

153 rows × 3 columns

Visualization

cluster_example = PyAGH.cluster(A)
plt.savefig('cluster_example.png', facecolor='w',dpi=300)

group=['1','1','1','1','1','1','1','2','2','2','2','2','2','2','2','2','2']
pca_example = PyAGH.pca(A,group=group)
pca_example.savefig('pca_example.png', facecolor='w',dpi=300)

heatmap_example = PyAGH.heat(A)
plt.savefig('heatmap_example.png', facecolor='w',dpi=300)

import graphviz
ped_selected = PyAGH.selectPed(data=ped,id=['17'],generation=3)
p = PyAGH.gragh(ped_selected)
graphviz.Source(p)

License

PyAGH is MIT licensed.