When we apply a dynamical mathematical model in a clinical setting for diagnosis and/or prognosis, we often need to estimate its parameters from a short time series observed for each patient. This estimation is the key for personalizing treatment options, but often accompanied with the uncertainty due to the shortness of the time series. Thus, here we review recent developments of our methods for estimating parameters and their uncertainty. We use a mathematical model of prostate cancer under intermittent androgen suppression for illustrating these methods. Using multiple methods will simultaneously make the path of precision medicine more solid.

Recent advances in next-generation sequencing technology allow high-throughput RNA sequencing (RNA-Seq) to be widely applied in transcriptomic studies. For model organisms with a reference genome, the first step in analysis of RNA-Seq data involves mapping of short-read sequences to the reference genome. Reference-guided transcriptome assembly is an optional step, which is recommended if the aim is to identify novel transcripts. Following read mapping, the primary interest of biologists in many RNA-Seq studies is the investigation of differential expression between experimental groups. In this review, we discuss recent developments in RNA-Seq data analysis applied to model organisms, including methods and algorithms for direct mapping, reference-guided transcriptome assembly and differential expression analysis, and provide insights for the future direction of RNA-Seq.

RNA-Seq is a revolutionary methodology that employs high-throughput sequencing technologies to enable highly sensitive detection and quantification of RNA in biological samples. Mapping of RNA-Seq data to a reference is a fundamental step for all forms of RNA-Seq data analysis in model organisms, and differential expression analysis is the primary interest of biologists in many RNA-Seq studies. In this review we discuss recent developments in these two fields and provide insights for the future direction of RNA-Seq. We see our review as a resource for the community that will enable researchers to select the most appropriate tools for RNA-Seq data analysis.

MicroRNA (miRNA) plays key roles in post-transcriptional regulations. Recently, a hypothesis of competing endogenous RNA (ceRNA) has been proposed as a new layer of gene regulation. Here, we revisit the common modelling framework and the current understanding of ceRNA effect. We propose that network topology could significantly influence it and the ceRNA effect at protein level could be much stronger than that at RNA level. We also provide a conditional independent binding equation to describe miRNA relative repression on different target, which could be applied to quantify siRNA off-target effect.

Molecule-based cancer subtyping from multi-omics data is one major goal of precision oncology. We summarize current unsupervised subtyping methods into three categories: direct integrative clustering, clustering of clusters and regulatory integrative clustering. Different categories use different strategies to deal with the heterogeneity and inter-dataset variations of multi-omics data. A few practical considerations on data pre-processing, post-clustering analysis and pathway-based analysis were also discussed.