Welcome to CropGBM!

Crop Genomic Breeding machine (CropGBM) is a multifunctional Python3 program that integrates data preprocessing, population structure analysis, SNP selection, phenotype prediction, and data visualization.It 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.

Chinese version documentation: https://ibreeding-ch.github.io

Download link: https://github.com/YuetongXU/CropGBM

Cite: 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.