B0965
Title: Contributions to mixture of distributions with application in sequenced data
Authors: Michelle de Klerk - University of Pretoria (South Africa) [presenting]
Frans Kanfer - University of Pretoria (South Africa)
Sollie Millard - University of Pretoria (South Africa)
Abstract: Mixture of distributions is an effective statistical method in the study of sequenced data as it considers all genes and identifies latent clusters within the data. Expression profiles are obtained using a negative binomial GLM. Identifying differentially expressed genes using a mixture modelling approach has shown to be favourable compared to traditional ANOVA like pairwise testing. An application to a sequenced dataset obtained from the Gene expression Omnibus is considered.
