Next maSigPro: updating maSigPro bioconductor package for RNA-seq time series

Abstract

Motivation: The widespread adoption of RNA-seq to quantitatively measure gene expression has increased the scope of sequencing experimental designs to include time-course experiments. maSigPro is an R package specifically suited for the analysis of time-course gene expression data, which was developed originally for microarrays and hence was limited in its application to count data.

Results: We have updated maSigPro to support RNA-seq time series analysis by introducing generalized linear models in the algorithm to support the modeling of count data while maintaining the traditional functionalities of the package. We show a good performance of the maSigPro-GLM method in several simulated time-course scenarios and in a real experimental dataset.

Availability and implementation: The package is freely available under the LGPL license from the Bioconductor Web site (http://bioconductor.org).

Fig. 1.

FDR and FNR for maSigPro-GLM at different levels of R² with 1 and 2 series

Fig. 2.

Random examples from genes selected with (A) maSigPro and edgeR, (B) maSigPro and not with edgeR, (C) with edgeR and not preselected with maSigPro and (D) with edgeR and not with maSigPro because R² < 0.5