MiPRIME: an integrated and intelligent platform for mining primer and probe sequences of microbial species

Zhiming Zhang1, Jing Ren1, Lili Ren2

  • 1Research and Development Department, Coyote Bioscience (Beijing) Co., Ltd., Building 22, Zone 3, Gaolizhang Road, Haidian District, Beijing, 10095, China.

Abstract

Insights

MiPRIME is a novel platform for extracting microbial primer and probe sequences from scientific literature. It uses a BioBERT model for high accuracy and a ranking score for efficient, species-specific primer recommendations.

Area of Science:

  • Microbiology
  • Bioinformatics
  • Computational Biology

Background:

  • Accurate detection of pathogenic microorganisms relies on effective primer and probe design.
  • Literature-derived primers are valuable but manual mining is time-consuming and limited in scope.

Purpose of the Study:

  • To develop an automated platform for mining microbial primer and probe sequences from scientific literature.
  • To facilitate high-frequency microbial gene discovery and user-defined primer design.

Main Methods:

  • Developed MiPRIME, a real-time Microbial Primer Mining platform.
  • Integrated over 40 million articles and 548,942 organisms.
  • Employed a BioBERT-based text mining model with 98.02% accuracy.
  • Implemented a primer ranking score (PRscore) for intelligent primer recommendation.

Main Results:

  • MiPRIME enables comprehensive microbial gene discovery from a global perspective.
  • The BioBERT model significantly reduces information processing time.
  • PRscore facilitates intelligent recommendation of species-specific primers, saving time and experimental costs.

Conclusions:

  • MiPRIME is a practical tool for primer mining in the pan-microbial field.
  • The platform advances microbial research by streamlining primer discovery.
  • MiPRIME offers a cost-effective solution for identifying effective primers and probes.