Huihui Li, Xin Li, Peng Zhang, Yingwei Feng, Junri Mi, Shang Gao, Lele Sheng, Mohsin Ali, Zikun Yang, Liang Li, Wei Fang, Wensheng Wang, Qian Qian, Fei Gu, Wenbin Zhou

Molecular Plant; 202; IF:27.50

DOI: 10.1016/j.molp.2024.03.002

Abstract:

In the era of big data and artificial intelligence, "smart breeding" has become a broad conceptual framework encompassing the paradigm shift of crop breeding to relying on analysis of high-throughput population genetics and phenomics data to conduct genomic selection, allowing identification and optimal use of the genetic potential in crop species (Sharma et al., 2022; Xiao et al., 2022; Xu et al., 2022). Most existing tools for analyzing high-throughput breeding data require extensive computational power, complex installation processes, and command-line expertise, and are therefore challenging and inconvenient for the majority of researchers and breeders (Brandies and Hogg, 2021). To overcome these limitations, we developed Smart Breeding Platform (https://sbp.ibreed.cn), a user-friendly, web-based tool for management and analysis of large-scale genetic, genomic, and phenomic data. This platform is freely accessible through the internet and allows users to import data, perform various statistical analyses, and conduct genome-wide association studies and genomic selection using both classical machine learning and deep-learning models. It will enable plant breeders to easily conduct the following steps: (1) efficiently record, manage, and process raw phenotypic and genotypic data; (2) perform phenotypic and population genetic analyses in highly customizable ways; and (3) easily conduct GWAS and genomic selection using classical machine learning and deep-learning models. Smart Breeding Platform contains four main sections (Figure 1): (1) Germplasm Data Management, (2) Test Management, (3) Genomic DataManagement, and (4) Data Analysis. Each section is described in detail below.