Toward multidimensional information: A derivatization-free UHPLC-QqQ MS/MS method for amino acid components of fingerprint.

Abstract

The analysis of fingerprint chemical composition is a meaningful way to excavate the multidimensional information of fingerprint, including the donor profiling information and the age of a fingerprint, which broadens the evidential values of fingerprint, especially for the partial and distorted fingerprint. But the research remains still in the pilot phases or is ongoing. Amino acids are the dominant organic substances in latent sweat fingerprint and influenced by many donor factors. Hence, their content reflects personal information of donors. Forensic science will be revolutionized if suspects can be individualized by their amino acid content. The diverse nature, distinct physicochemical properties, and ultra-micro levels of amino acids present in fingerprints make it hard to detect. A high sensitivity method for detecting and quantifying multiple amino acid components is required. UHPLC-QqQ MS/MS offers high sensitivity, high separation, simultaneous multicomponents detection, and no derivatization, making it an ideal method for detecting and analyzing amino acids in fingerprints. Therefore, in this study, we propose and validate an efficient UHPLC-QqQ MS/MS method for the extraction and analysis of 13 amino acids from fingerprint. We compared the results of amino acids of 10 different substrates and found that the inherent amino acids in most porous substrates would have been extracted along with the fingerprint amino acids, making them unsuitable for quantitative amino acid analysis. Instead, plastic sheets are ideal substrates for laboratory studies. Then, extensive experiments were conducted among 30 donors for multidimensional information analysis. The type of samples analyzed were eccrine-rich fingerprints. A Binary Logistic Regression (BLR) model was developed, and the female and male donors were successfully differentiated by amino acids in fingerprints. Two other mathematical models were also developed to verify the accuracy, and all three different mathematical models were able to identify donors of different genders with over 90% accuracy. This demonstrates that amino acids have the potential to provide more information for donors as metabolic markers. In the future, we will conduct a series of experiments to analyze more multidimensional information for individual identification by amino acid content in the fingerprint.