How many biological replicates are needed in an RNA-seq experiment and which differential expression tool should you use?

RNA-seq is now the technology of choice for genome-wide differential gene expression experiments, but it is not clear how many biological replicates are needed to ensure valid biological interpretation of the results or which statistical tools are best for analyzing the data. An RNA-seq experiment with 48 biological replicates in each of two conditions was performed to answer these questions and provide guidelines for experimental design. With three biological replicates, nine of the 11 tools evaluated found only 20%-40% of the significantly differentially expressed (SDE) genes identified with the full set of 42 clean replicates. This rises to >85% for the subset of SDE genes changing in expression by more than fourfold. To achieve >85% for all SDE genes regardless of fold change requires more than 20 biological replicates. The same nine tools successfully control their false discovery rate at ≲5% for all numbers of replicates, while the remaining two tools fail to control their FDR adequately, particularly for low numbers of replicates. For future RNA-seq experiments, these results suggest that at least six biological replicates should be used, rising to at least 12 when it is important to identify SDE genes for all fold changes. If fewer than 12 replicates are used, a superior combination of true positive and false positive performances makes edgeR and DESeq2 the leading tools. For higher replicate numbers, minimizing false positives is more important and DESeq marginally outperforms the other tools.

Statistical models for RNA-seq data derived from a two-condition 48-replicate experiment.

MOTIVATION: High-throughput RNA sequencing (RNA-seq) is now the standard method to determine differential gene expression. Identifying differentially expressed genes crucially depends on estimates of read-count variability. These estimates are typically based on statistical models such as the negative binomial distribution, which is employed by the tools edgeR, DESeq and cuffdiff. Until now, the validity of these models has usually been tested on either low-replicate RNA-seq data or simulations. RESULTS: A 48-replicate RNA-seq experiment in yeast was performed and data tested against theoretical models. The observed gene read counts were consistent with both log-normal and negative binomial distributions, while the mean-variance relation followed the line of constant dispersion parameter of ∼0.01. The high-replicate data also allowed for strict quality control and screening of 'bad' replicates, which can drastically affect the gene read-count distribution. AVAILABILITY AND IMPLEMENTATION: RNA-seq data have been submitted to ENA archive with project ID PRJEB5348. CONTACT: g.j.barton@dundee.ac.uk.

Filaggrin-stratified transcriptomic analysis of pediatric skin identifies mechanistic pathways in patients with atopic dermatitis.

BACKGROUND: Atopic dermatitis (AD; eczema) is characterized by a widespread abnormality in cutaneous barrier function and propensity to inflammation. Filaggrin is a multifunctional protein and plays a key role in skin barrier formation. Loss-of-function mutations in the gene encoding filaggrin (FLG) are a highly significant risk factor for atopic disease, but the molecular mechanisms leading to dermatitis remain unclear. OBJECTIVE: We sought to interrogate tissue-specific variations in the expressed genome in the skin of children with AD and to investigate underlying pathomechanisms in atopic skin. METHODS: We applied single-molecule direct RNA sequencing to analyze the whole transcriptome using minimal tissue samples. Uninvolved skin biopsy specimens from 26 pediatric patients with AD were compared with site-matched samples from 10 nonatopic teenage control subjects. Cases and control subjects were screened for FLG genotype to stratify the data set. RESULTS: Two thousand four hundred thirty differentially expressed genes (false discovery rate, P < .05) were identified, of which 211 were significantly upregulated and 490 downregulated by greater than 2-fold. Gene ontology terms for "extracellular space" and "defense response" were enriched, whereas "lipid metabolic processes" were downregulated. The subset of FLG wild-type cases showed dysregulation of genes involved with lipid metabolism, whereas filaggrin haploinsufficiency affected global gene expression and was characterized by a type 1 interferon-mediated stress response. CONCLUSION: These analyses demonstrate the importance of extracellular space and lipid metabolism in atopic skin pathology independent of FLG genotype, whereas an aberrant defense response is seen in subjects with FLG mutations. Genotype stratification of the large data set has facilitated functional interpretation and might guide future therapy development.

Re-replication induced by geminin depletion occurs from G2 and is enhanced by checkpoint activation.

To prevent re-replication of DNA in a single cell cycle, the licensing of replication origins by Mcm2-7 is prevented during S and G2 phases. Animal cells achieve this by cell-cycle-regulated proteolysis of the essential licensing factor Cdt1 and inhibition of Cdt1 by geminin. Here we investigate the consequences of ablating geminin in synchronised human U2OS cells. Following geminin loss, cells complete an apparently normal S phase, but a proportion arrest at the G2-M boundary. When Cdt1 accumulates in these cells, DNA re-replicates, suggesting that the key role of geminin is to prevent re-licensing in G2. If cell cycle checkpoints are inhibited in cells lacking geminin, cells progress through mitosis and less re-replication occurs. Checkpoint kinases thereby amplify re-replication into an all-or-nothing response by delaying geminin-depleted cells in G2. Deep DNA sequencing revealed no preferential re-replication of specific genomic regions after geminin depletion. This is consistent with the observation that cells in G2 have lost their replication timing information. By contrast, when Cdt1 is overexpressed or is stabilised by the neddylation inhibitor MLN4924, re-replication can occur throughout S phase.

Active and adaptive plasticity in a changing climate.

Better understanding of the mechanistic basis of plant plasticity will enhance efforts to breed crops resilient to predicted climate change. However, complexity in plasticity's conceptualisation and measurement may hinder fruitful crossover of concepts between disciplines that would enable such advances. We argue active adaptive plasticity is particularly important in shaping the fitness of wild plants, representing the first line of a plant's defence to environmental change. Here, we define how this concept may be applied to crop breeding, suggest appropriate approaches to measure it in crops, and propose a refocussing on active adaptive plasticity to enhance crop resilience. We also discuss how the same concept may have wider utility, such as in ex situ plant conservation and reintroductions.

Nanopore direct RNA sequencing maps the complexity of Arabidopsis mRNA processing and m6A modification.

Understanding genome organization and gene regulation requires insight into RNA transcription, processing and modification. We adapted nanopore direct RNA sequencing to examine RNA from a wild-type accession of the model plant Arabidopsis thaliana and a mutant defective in mRNA methylation (m6A). Here we show that m6A can be mapped in full-length mRNAs transcriptome-wide and reveal the combinatorial diversity of cap-associated transcription start sites, splicing events, poly(A) site choice and poly(A) tail length. Loss of m6A from 3' untranslated regions is associated with decreased relative transcript abundance and defective RNA 3' end formation. A functional consequence of disrupted m6A is a lengthening of the circadian period. We conclude that nanopore direct RNA sequencing can reveal the complexity of mRNA processing and modification in full-length single molecule reads. These findings can refine Arabidopsis genome annotation. Further, applying this approach to less well-studied species could transform our understanding of what their genomes encode.

Relative Abundance of Transcripts ( RATs): Identifying differential isoform abundance from RNA-seq.

The biological importance of changes in RNA expression is reflected by the wide variety of tools available to characterise these changes from RNA-seq data. Several tools exist for detecting differential transcript isoform usage (DTU) from aligned or assembled RNA-seq data, but few exist for DTU detection from alignment-free RNA-seq quantifications. We present the RATs, an R package that identifies DTU transcriptome-wide directly from transcript abundance estimates. RATs is unique in applying bootstrapping to estimate the reliability of detected DTU events and shows good performance at all replication levels (median false positive fraction < 0.05). We compare RATs to two existing DTU tools, DRIM-Seq & SUPPA2, using two publicly available simulated RNA-seq datasets and a published human RNA-seq dataset, in which 248 genes have been previously identified as displaying significant DTU. RATs with default threshold values on the simulated Human data has a sensitivity of 0.55, a Matthews correlation coefficient of 0.71 and a false discovery rate (FDR) of 0.04, outperforming both other tools. Applying the same thresholds for SUPPA2 results in a higher sensitivity (0.61) but poorer FDR performance (0.33). RATs and DRIM-seq use different methods for measuring DTU effect-sizes complicating the comparison of results between these tools, however, for a likelihood-ratio threshold of 30, DRIM-Seq has similar FDR performance to RATs (0.06), but worse sensitivity (0.47). These differences persist for the simulated drosophila dataset. On the published human RNA-seq dataset the greatest agreement between the tools tested is 53%, observed between RATs and SUPPA2. The bootstrapping quality filter in RATs is responsible for removing the majority of DTU events called by SUPPA2 that are not reported by RATs. All methods, including the previously published qRT-PCR of three of the 248 detected DTU events, were found to be sensitive to annotation differences between Ensembl v60 and v87.

Improved annotation of 3' untranslated regions and complex loci by combination of strand-specific direct RNA sequencing, RNA-Seq and ESTs.

The reference annotations made for a genome sequence provide the framework for all subsequent analyses of the genome. Correct and complete annotation in addition to the underlying genomic sequence is particularly important when interpreting the results of RNA-seq experiments where short sequence reads are mapped against the genome and assigned to genes according to the annotation. Inconsistencies in annotations between the reference and the experimental system can lead to incorrect interpretation of the effect on RNA expression of an experimental treatment or mutation in the system under study. Until recently, the genome-wide annotation of 3' untranslated regions received less attention than coding regions and the delineation of intron/exon boundaries. In this paper, data produced for samples in Human, Chicken and A. thaliana by the novel single-molecule, strand-specific, Direct RNA Sequencing technology from Helicos Biosciences which locates 3' polyadenylation sites to within +/- 2 nt, were combined with archival EST and RNA-Seq data. Nine examples are illustrated where this combination of data allowed: (1) gene and 3' UTR re-annotation (including extension of one 3' UTR by 5.9 kb); (2) disentangling of gene expression in complex regions; (3) clearer interpretation of small RNA expression and (4) identification of novel genes. While the specific examples displayed here may become obsolete as genome sequences and their annotations are refined, the principles laid out in this paper will be of general use both to those annotating genomes and those seeking to interpret existing publically available annotations in the context of their own experimental data.