Density-weighted Algorithms for Similarity Computation and Cluster Tree Construction in the RAPD Analysis of Natural Cordyceps sinensis

¹ Department of Mechanical and Aerospace Engineering, University of California San Diego, La Jolla, CA, USA

² Pharmanex Beijing Pharmacology Center, Beijing 100088, China

³ Department of Applied Biology and Chemistry Technology, Hong Kong Polytechnic University, Hong Kong

⁴ NS Center for Anti-aging Research, Provo, UT 84601, USA

^# These authors (LN and YSY) contributed equally to this work

*Corresponding author:

Dr. J-S Zhu

NS Center for Anti-aging Research

Provo, UT 84601

USA

Tel: +1(858) 705-3790

Fax: +1(858) 777-5435

E-mail: zhujosh@gmail.com

Abstract

Objective: The goal of this study was to develop and validate density-weighted arithmetic methods for the analysis of the dynamic maturational changes in random amplified polymorphic DNA (RAPD) polymorphisms in Cordyceps sinensis containing multiple fungi. Methods: Ten random primers were used for PCR amplification to monitor changes in the RAPD molecular marker polymorphisms in caterpillar body, stroma and ascocarp portion samples of C. sinensis collected at different stages of maturation. We compared (1) the density-unweighted Nei-Li similarity equation and new similarity equations considering the densities of DNA bands and (2) the density-unweighted and weighted arithmetic methods for cluster analysis. Results: The algorithm using the Nei-Li similarity equation did not account for the differences in the density of the matched randomly amplified amplicons, whereas the new similarity equations were capable of integrating different amplicon densities into the similarity computation. With improvements in the similarity computation and cluster construction, the new methods revealed dynamic maturational changes in the RAPD polymorphisms of molecular markers in C. sinensis caterpillar bodies and stromata. The polymorphism analysis revealed similarities of 74%-88% between the RAPD polymorphisms of the ascocarp portion versus premature and mature C. sinensis stromata, but low similarities of<70% versus the caterpillar bodies and maturing stromata. Apparent dissimilarities (similarity <66%) were also found between Hirsutella sinensis and C. sinensis samples. Conclusions: These new algorithmic methods represent an advance in similarity computations when comparing polymorphisms of RAPD molecular markers in C. sinensis and provide accurate analytic tools to capture the molecular information related to the dynamic changes in differential expression of C. sinensis intrinsic fungi during maturation. The apparent dissimilarity in RAPD polymorphisms between H. sinensis and C. sinensis suggests differences of mycological background and challenges the hypothesis of the anamorph-teleomorph connection between H. sinensis and Ophiocordyceps sinensis.

Keywords: RAPD (Random Amplified Polymorphic DNA) polymorphism of molecular markers; density-weighted similarity computation; density-weighted cluster construction; Cordyceps sinensis; Ophiocordyceps sinensis; Hirsutella sinensis.

Download the full article (PDF)