A probabilistic approach to heroin signatures.

Area of Science:

• Forensic Chemistry
• Chemical Profiling
• Drug Analysis

Background:

• Heroin seizures provide crucial intelligence on illicit drug supply chains.
• Determining the geographical origin of heroin is vital for law enforcement and policy.
• Existing methods for origin determination have limitations in accuracy and refinement.

Purpose of the Study:

• To develop a probabilistic method for determining the geographical origin of seized heroin.
• To apply Bayes Theorem using compound ratios for more accurate origin assignment.
• To compare the effectiveness of this new method against traditional ratio approaches.

Main Methods:

• Analysis of compound ratios (e.g., codeine/morphine, noscapine/morphine) in 2549 heroin seizures.
• Calculation of probability density functions for these ratios based on known world regions.
• Application of Bayes Theorem with prior probabilities to estimate the likelihood of a sample originating from a specific region.

Main Results:

• The developed method allows for the calculation of likelihood ratios for unknown heroin samples.
• Probabilistic assignment of geographical origin based on compound profiles was achieved.
• Results demonstrated superior refinement in geographical intelligence compared to simple ratio methods.

Conclusions:

• The Bayesian approach using compound ratios offers a more sophisticated tool for heroin origin determination.
• This method can be implemented using readily available tools like spreadsheets.
• Enhanced intelligence on drug origins can aid in combating illicit drug trafficking.