Genetic diversity and core collection of Polygonati Rhizoma in China via SSR markers

TL;DR

This study uses genetic markers to analyze the diversity of Polygonati Rhizoma in China and creates a core collection for conservation and improvement.

Contribution

The study introduces a machine learning-based method for species discrimination and constructs a core collection for Polygonatum species.

Findings

Machine learning models achieved over 81% accuracy in identifying Polygonatum species.

Significant genetic diversity was observed among the four Polygonati Rhizoma species.

A core collection with 84.59% allele retention was developed to minimize genetic redundancy.

Abstract

Polygonati Rhizoma is a Traditional Chinese medicine derived from three species of the Polygonatum Mill. (Polygonatum kingianum, Polygonatum sibiricum, and Polygonatum cyrtonema), renowned for its culinary and medicinal uses. Despite extensive research, comprehensive studies on the population genetic diversity and core collection construction of medicinal Polygonatum species remain scarce.

To address this, we employed 18 highly polymorphic SSR markers to develop two machine learning models for species discrimination. Subsequently, we performed comprehensive population genetic analyses on 175 accessions, followed by core germplasm construction.

The study demonstrated that the machine learning-based approach achieved consistently high discrimination accuracy, exceeding 81% for Polygonatum species identification. Among the four investigated Polygonati Rhizoma, significant variations in genetic diversity were observed. Cluster and population structure analyses identified three primary subgroups. A core collection was constructed through stepwise clustering based on genetic distance. The C78 primary core collection achieved an allele retention rate of 84.59%, with minimal genetic redundancy.

These findings provide a robust foundation for the conservation of medicinal Polygonatum spp. germplasm and offer potential resources for future genetic improvement and variety selection.

Figures