Computer assisted substance identification in systematic toxicological analysis: New life for old methods?
Rafael Linden1, Lilian L Feltraco, Luana Christine Comerlato
1Instituto de Ciências da Saúde, Universidade Feevale, Rodovia RS 239 n. 2755, CEP 93352-000, Novo Hamburgo, RS, Brazil. rafael.linden@feevale.br
Forensic Science International
|June 18, 2010
Summary
A new software aids toxicological analysis by processing analytical data for substance identification. This tool offers a cost-effective, probabilistic approach for resource-limited labs, improving accuracy in clinical and forensic toxicology.
Area of Science:
- Toxicological analysis
- Analytical chemistry
- Forensic science
Background:
- Systematic toxicological analysis is crucial in clinical and forensic settings.
- Mass spectrometry is the standard but can be resource-intensive.
- Resource-limited laboratories require alternative identification methods.
Purpose of the Study:
- To develop an Internet-based software for toxicological data processing.
- To provide a probabilistic approach for substance identification.
- To create a valuable alternative for resource-limited laboratories.
Main Methods:
- Developed Internet-based software for processing analytical data.
- Utilized a probabilistic approach with a database of retention indices (GC) and relative retention times/UV absorption maxima (HPLC).
- Ranked candidate substances using a Similarity Index based on standardized analytical conditions.
Main Results:
- Achieved a Mean List Length of 1.6 with a database of 1762 compounds.
- Correctly identified the substance in 34 out of 35 spiked samples at the highest Similarity Index.
- Demonstrated comparable performance to more sophisticated methods.
Conclusions:
- The developed software provides accurate substance identification in toxicology.
- It is a valuable, cost-effective tool for resource-limited clinical and forensic laboratories.
- The probabilistic approach enhances the reliability of toxicological analysis.
