06/07/2017 - Identifying biomarkers for progesterone using mass spectrometry data - Palestrante: Nancy L. Garcia (UNICAMP)

Seminário de Probabilidade e Sistemas Complexos

Palestrante: Nancy L. Garcia (IMECC-UNICAMP)

Local: Sala 38 do Departamento de Estatística (DEs) da UFSCar.

Data e horário: Quinta feira, dia 6 de julho de 2017 às 14h.

Título: Identifying biomarkers for progesterone using mass spectrometry data.

Resumo: Changes in the lipid composition of the bovine uterus exposed to greater (LF-LCL group) or lower (SF-SCL group) concentrations of progesterone during post-ovulation were investigated by matrix assisted laser desorption ionization-mass spectrometry (MALDI-MS). For each cow two measurements were made and after preprocessing of the data, 76 m/z values were selected identifying specific ions in the spectra. Due to the small sample size, 7 cows in LF-LCL group and 10 cows in SF-SCL group, the usual PCA methods could not identify biomarkers that discriminated between groups. A model-based approach was therefore proposed and found able to classify the MALDI-MS data, fitting a mixture of bivariate beta distributions to accommodate the large number of zero observations as well as the bivariate nature of the data.