How well do RNA-Seq differential gene expression tools perform in a complex eukaryote? A case study in Arabidopsis thaliana.

MOTIVATION: RNA-seq experiments are usually carried out in three or fewer replicates. In order to work well with so few samples, differential gene expression (DGE) tools typically assume the form of the underlying gene expression distribution. In this paper, the statistical properties of gene expression from RNA-seq are investigated in the complex eukaryote, Arabidopsis thaliana, extending and generalizing the results of previous work in the simple eukaryote Saccharomyces cerevisiae. RESULTS: We show that, consistent with the results in S.cerevisiae, more gene expression measurements in A.thaliana are consistent with being drawn from an underlying negative binomial distribution than either a log-normal distribution or a normal distribution, and that the size and complexity of the A.thaliana transcriptome does not influence the false positive rate performance of nine widely used DGE tools tested here. We therefore recommend the use of DGE tools that are based on the negative binomial distribution. AVAILABILITY AND IMPLEMENTATION: The raw data for the 17 WT Arabidopsis thaliana datasets is available from the European Nucleotide Archive (E-MTAB-5446). The processed and aligned data can be visualized in context using IGB (Freese et al., 2016), or downloaded directly, using our publicly available IGB quickload server at https://compbio.lifesci.dundee.ac.uk/arabidopsisQuickload/public_quickload/ under 'RNAseq>Froussios2019'. All scripts and commands are available from github at https://github.com/bartongroup/KF_arabidopsis-GRNA. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Major transcriptome re-organisation and abrupt changes in signalling, cell cycle and chromatin regulation at neural differentiation in vivo.

Here, we exploit the spatial separation of temporal events of neural differentiation in the elongating chick body axis to provide the first analysis of transcriptome change in progressively more differentiated neural cell populations in vivo. Microarray data, validated against direct RNA sequencing, identified: (1) a gene cohort characteristic of the multi-potent stem zone epiblast, which contains neuro-mesodermal progenitors that progressively generate the spinal cord; (2) a major transcriptome re-organisation as cells then adopt a neural fate; and (3) increasing diversity as neural patterning and neuron production begin. Focussing on the transition from multi-potent to neural state cells, we capture changes in major signalling pathways, uncover novel Wnt and Notch signalling dynamics, and implicate new pathways (mevalonate pathway/steroid biogenesis and TGFß). This analysis further predicts changes in cellular processes, cell cycle, RNA-processing and protein turnover as cells acquire neural fate. We show that these changes are conserved across species and provide biological evidence for reduced proteasome efficiency and a novel lengthening of S phase. This latter step may provide time for epigenetic events to mediate large-scale transcriptome re-organisation; consistent with this, we uncover simultaneous downregulation of major chromatin modifiers as the neural programme is established. We further demonstrate that transcription of one such gene, HDAC1, is dependent on FGF signalling, making a novel link between signals that control neural differentiation and transcription of a core regulator of chromatin organisation. Our work implicates new signalling pathways and dynamics, cellular processes and epigenetic modifiers in neural differentiation in vivo, identifying multiple new potential cellular and molecular mechanisms that direct differentiation.

The impact of KLF2 modulation on the transcriptional program and function of CD8 T cells.

Krüppel-like factor 2 (KLF2) is a transcription factor that is highly expressed in quiescent T lymphocytes and downregulated in effector T cells. We now show that antigen receptor engagement downregulates KLF2 expression in a graded response determined by the affinity of T cell antigen receptor (TCR) ligand and the integrated activation of protein kinase B and the MAP kinases ERK1/2. The present study explores the importance of KLF2 downregulation and reveals that the loss of KLF2 controls a select portion of the CD8 effector T cell transcriptional program. In particular, KLF2 loss is required for CD8 T cells to express the inflammatory chemokine receptor CXCR3 and for maximum clonal expansion of T cells. KLF2 thus negatively controls the ability of CD8 T cells to respond to the CXCR3 ligand CXCL10. Strikingly, the KLF2 threshold for restraining expression of CXCR3 is very low and quite distinct to the KLF2 threshold for restraining T cell proliferation. KLF2 is thus an analogue (tunable) not a digital (on/off) cellular switch where the magnitude of KLF2 expression differentially modifies the T cell responses.

PTEN protein phosphatase activity correlates with control of gene expression and invasion, a tumor-suppressing phenotype, but not with AKT activity.

The tumor suppressor phosphatase and tensin homolog deleted on chromosome 10 (PTEN) has a well-characterized lipid phosphatase activity and a poorly characterized protein phosphatase activity. We show that both activities are required for PTEN to inhibit cellular invasion and to mediate most of its largest effects on gene expression. PTEN appears to dephosphorylate itself at threonine 366, and mutation of this site makes lipid phosphatase activity sufficient for PTEN to inhibit invasion. We propose that the dominant role for PTEN's protein phosphatase activity is autodephosphorylation-mediated regulation of its lipid phosphatase activity. Because PTEN's regulation of invasion and these changes in gene expression required lipid phosphatase activity, but did not correlate with the total cellular abundance of its phosphatidylinositol 3,4,5-trisphosphate (PIP3) lipid substrate or AKT activity, we propose that localized PIP3 signaling may play a role in those PTEN-mediated processes that depend on both its protein and lipid phosphatase activities. Finally, we identified a tumor-derived PTEN mutant selectively lacking protein phosphatase activity, indicating that in some circumstances the regulation of invasion and not that of AKT can correlate with PTEN-mediated tumor suppression.