By measuring gene expression at an unprecedented resolution and throughput, RNA-seq has played a pivotal
role in studying biological functions. Its typical application in clinical medicine is to identify the discrepancies of gene
expression between two different types of cancer cells, sensitive and resistant to chemotherapeutic treatment, in a hope to
predict drug response. Here we modified and used a mechanistic model to identify distinct patterns of gene expression in
response of different types of breast cancer cell lines to chemotherapeutic treatment. This model was founded on a mixture
likelihood of Poisson-distributed transcript read data, with each mixture component specified by the Skellam function. By
estimating and comparing the amount of gene expression in each environment, the model can test how genes alter their
expression in response to environment and how different genes interact with each other in the responsive process. Using
the modified model, we identified the alternations of gene expression between two cell lines of breast cancer, resistant and
sensitive to tamoxifen, which allows us to interpret the expression mechanism of how genes respond to metabolic differences
between the two cell types. The model can have a general implication for studying the plastic pattern of gene expression
across different environments measured by RNA-seq.
Keywords: RNA-seq, Phenotypic plasticity, Gene-environment interaction, Clustering.
Rights & PermissionsPrintExport