LC-MS estimation model developed for biological sample analysis

A pair of scientists from the Rollins School of Public Health at Emory University, Atlanta, have developed a new statistical method for liquid chromatography-mass spectrometry (LC-MS) analysis.

The team, Tianwei Yu and Hesen Peng, explain that peak modelling represents a core component in pre-processing data for LC-MS studies.

In turn, LC-MS offers a way to quantify metabolites, allowing complex biological samples to be scrutinised.

The pair have now developed a new estimation model by carrying out an extensive series of simulations accompanied by testing on real-world samples.

In their report, they write: "To accurately quantify partially overlapping peaks, we developed a deconvolution method using the bi-Gaussian mixture model combined with statistical model selection."

A bi-Gaussian function also forms the core of their process to estimate and quantify asymmetric peaks when there are high levels of noise in the sample.

BMC Bioinformatics carries research and development of new statistical models and processes for the analysis of biological data in all forms.