Crop Genomic Breeding machine (CropGBM) is a multifunctional Python3 program that integrates data preprocessing, population structure analysis, SNP selection, phenotype prediction, and data visualization. Has the following advantages:
- Use LightGBM algorithm to quickly and accurately predict phenotype values and support GPU-accelerated training.
- Supports selection and visualization of SNPs that are strongly related to phenotype.
- Support PCA and t-SNE two dimensionality reduction algorithms to extract SNP information.
- Support Kmeans and OPTICS two clustering algorithms to analyze the sample population structure.
- Plot histograms of heterozygosity rate, deletion rate, and frequency of alleles for genotype data.
English version documentation: https://ibreeding.github.io
Chinese version documentation: https://ibreeding-ch.github.io
Download source code : https://github.com/YuetongXU/CropGBM/releases/tag/cropgbm-v1.1.2
$ conda install -c xu_cau_cab cropgbm
$ pip install --user cropgbm
$ tar -zxf CropGBM.tar.gz
# Install Python package dependencies of CropGBM: setuptools, wheel, numpy, scipy, pandas, scikit-learn, lightgbm, matplotlib, seaborn
$ pip install --user setuptools wheel numpy scipy pandas scikit-learn lightgbm matplotlib seaborn
# Install external dependencies of CropGBM: PLINK 1.90
$ wget s3.amazonaws.com/plink1-assets/plink_linux_x86_64_20191028.zip
$ mkdir plink_1.90
$ unzip plink_linux_x86_64_20191028.zip -d ./plink_1.90
# Add CropGBM, PLINK to the system environment variables for quick use:
$ vi ~/.bashrc
export PATH="/userpath/CropGBM:$PATH"
export PATH="/userpath/plink1.90:$PATH"
$ source ~/.bashrc
Enter the ‘/miniconda3/pkgs/cropgbm-1.1.2-py39_0/info/test’ folder
Run the run_test.py to check whether cropgbm can run successfully locally.
Citation: Jun Yan, Yuetong Xu, Qian Cheng, Shuqin Jiang, Qian Wang, Yingjie Xiao, Chuang Ma, Jianbing Yan and Xiangfeng Wang. LightGBM: accelerated genomically-designed crop breeding through ensemble learning.
Supplementary Information: Support data and materials for the manuscript is available at https://github.com/YuetongXU/Cropgbm-Paper
Contact us: cropgbm@163.com
Note: Academic users can download directly, industrial users first contact us.