Utilizing Big Data to Identify Tiny Toxic Components: Digitalis

Area of Science:

• Genomics
• Bioinformatics
• Pharmacognosy

Background:

• The botanical genus *Digitalis* contains toxic and medicinal compounds with a narrow therapeutic range.
• Contamination of dietary supplements with *Digitalis* species, like *Digitalis lanata*, poses a significant health risk.
• Previous detection methods relied on chemical analysis, which can be time-consuming.

Purpose of the Study:

• To evaluate the efficacy of shallow genome sequencing strategies for rapid detection of *Digitalis* contamination in dietary supplements.
• To develop and validate a genomic approach for identifying *Digitalis* at both genus and species levels.

Main Methods:

• Utilized a modified Site Identification from Short Read Sequences (SISRS) bioinformatics pipeline with whole-genome sequence data.
• Generated genus-level and species-informative single nucleotide polymorphisms (SNPs).
• Simulated contamination by spiking varying percentages (0-10%) of *Digitalis* whole-genome sequence data into common dietary supplement ingredients.

Main Results:

• Successfully generated over 2 million genus-level SNPs and species-informative SNPs.
• Reliably detected *Digitalis* at the genus level in simulated contamination scenarios.
• Effectively discriminated between different *Digitalis* species, even at low contamination levels.

Conclusions:

• Shallow genome sequencing is a rapid and reliable method for detecting toxic botanical species like *Digitalis* in food products and dietary supplements.
• Genomic methods offer a powerful alternative to traditional chemical analyses for ensuring product safety.
• This work provides a framework for developing novel DNA-based assays for botanical contaminant detection.