Artificial intelligence-assisted automated heart failure detection and classification from electronic health records.

AIMS: Electronic health records (EHR) linked to Digital Imaging and Communications in Medicine (DICOM), biological specimens, and deep learning (DL) algorithms could potentially improve patient care through automated case detection and surveillance. We hypothesized that by applying keyword searches to routinely stored EHR, in conjunction with AI-powered automated reading of DICOM echocardiography images and analysing biomarkers from routinely stored plasma samples, we were able to identify heart failure (HF) patients. METHODS AND RESULTS: We used EHR data between 1993 and 2021 from Tayside and Fife (~20% of the Scottish population). We implemented a keyword search strategy complemented by filtering based on International Classification of Diseases (ICD) codes and prescription data to EHR data set. We then applied DL for the automated interpretation of echocardiographic DICOM images. These methods were then integrated with the analysis of routinely stored plasma samples to identify and categorize patients into HF with reduced ejection fraction (HFrEF), HF with preserved ejection fraction (HFpEF), and controls without HF. The final diagnosis was verified through a manual review of medical records, measured natriuretic peptides in stored blood samples, and by comparing clinical outcomes among groups. In our study, we selected the patient cohort through an algorithmic workflow. This process started with 60 850 EHR data and resulted in a final cohort of 578 patients, divided into 186 controls, 236 with HFpEF, and 156 with HFrEF, after excluding individuals with mismatched data or significant valvular heart disease. The analysis of baseline characteristics revealed that compared with controls, patients with HFrEF and HFpEF were generally older, had higher BMI, and showed a greater prevalence of co-morbidities such as diabetes, COPD, and CKD. Echocardiographic analysis, enhanced by DL, provided high coverage, and detailed insights into cardiac function, showing significant differences in parameters such as left ventricular diameter, ejection fraction, and myocardial strain among the groups. Clinical outcomes highlighted a higher risk of hospitalization and mortality for HF patients compared with controls, with particularly elevated risk ratios for both HFrEF and HFpEF groups. The concordance between the algorithmic selection of patients and manual validation demonstrated high accuracy, supporting the effectiveness of our approach in identifying and classifying HF subtypes, which could significantly impact future HF diagnosis and management strategies. CONCLUSIONS: Our study highlights the feasibility of combining keyword searches in EHR, DL automated echocardiographic interpretation, and biobank resources to identify HF subtypes.

PERK signaling promotes mitochondrial elongation by remodeling membrane phosphatidic acid.

Endoplasmic reticulum (ER) stress and mitochondrial dysfunction are linked in the onset and pathogenesis of numerous diseases. This has led to considerable interest in defining the mechanisms responsible for regulating mitochondria during ER stress. The PERK signaling arm of the unfolded protein response (UPR) has emerged as a prominent ER stress-responsive signaling pathway that regulates diverse aspects of mitochondrial biology. Here, we show that PERK activity promotes adaptive remodeling of mitochondrial membrane phosphatidic acid (PA) to induce protective mitochondrial elongation during acute ER stress. We find that PERK activity is required for ER stress-dependent increases in both cellular PA and YME1L-dependent degradation of the intramitochondrial PA transporter PRELID1. These two processes lead to the accumulation of PA on the outer mitochondrial membrane where it can induce mitochondrial elongation by inhibiting mitochondrial fission. Our results establish a new role for PERK in the adaptive remodeling of mitochondrial phospholipids and demonstrate that PERK-dependent PA regulation adapts organellar shape in response to ER stress.

The protein kinases of Dictyostelia and their incorporation into a signalome.

Protein kinases are major regulators of cellular processes, but the roles of most kinases remain unresolved. Dictyostelid social amoebas have been useful in identifying functions for 30% of its kinases in cell migration, cytokinesis, vesicle trafficking, gene regulation and other processes but their upstream regulators and downstream effectors are mostly unknown. Comparative genomics can assist to distinguish between genes involved in deeply conserved core processes and those involved in species-specific innovations, while co-expression of genes as evident from comparative transcriptomics can provide cues to the protein complement of regulatory networks. Genomes and developmental and cell-type specific transcriptomes are available for species that span the 0.5 billion years of evolution of Dictyostelia from their unicellular ancestors. In this work we analysed conservation and change in the abundance, functional domain architecture and developmental regulation of protein kinases across the 4 major taxon groups of Dictyostelia. All data are summarized in annotated phylogenetic trees of the kinase subtypes and accompanied by functional information of all kinases that were experimentally studied. We detected 393 different protein kinase domains across the five studied genomes, of which 212 were fully conserved. Conservation was highest (71%) in the previously defined AGC, CAMK, CK1, CMCG, STE and TKL groups and lowest (26%) in the "other" group of typical protein kinases. This was mostly due to species-specific single gene amplification of "other" kinases. Apart from the AFK and α-kinases, the atypical protein kinases, such as the PIKK and histidine kinases were also almost fully conserved. The phylogeny-wide developmental and cell-type specific expression profiles of the protein kinase genes were combined with profiles from the same transcriptomic experiments for the families of G-protein coupled receptors, small GTPases and their GEFs and GAPs, the transcription factors and for all genes that upon lesion generate a developmental defect. This dataset was subjected to hierarchical clustering to identify clusters of co-expressed genes that potentially act together in a signalling network. The work provides a valuable resource that allows researchers to identify protein kinases and other regulatory proteins that are likely to act as intermediates in a network of interest.

A Comprehensive Enumeration of the Human Proteostasis Network. 2. Components of the Autophagy-Lysosome Pathway.

The condition of having a healthy, functional proteome is known as protein homeostasis, or proteostasis. Establishing and maintaining proteostasis is the province of the proteostasis network, approximately 2,700 components that regulate protein synthesis, folding, localization, and degradation. The proteostasis network is a fundamental entity in biology that is essential for cellular health and has direct relevance to many diseases of protein conformation. However, it is not well defined or annotated, which hinders its functional characterization in health and disease. In this series of manuscripts, we aim to operationally define the human proteostasis network by providing a comprehensive, annotated list of its components. We provided in a previous manuscript a list of chaperones and folding enzymes as well as the components that make up the machineries for protein synthesis, protein trafficking into and out of organelles, and organelle-specific degradation pathways. Here, we provide a curated list of 838 unique high-confidence components of the autophagy-lysosome pathway, one of the two major protein degradation systems in human cells.

Machine learning models in trusted research environments - understanding operational risks.

Introduction: Trusted research environments (TREs) provide secure access to very sensitive data for research. All TREs operate manual checks on outputs to ensure there is no residual disclosure risk. Machine learning (ML) models require very large amount of data; if this data is personal, the TRE is a well-established data management solution. However, ML models present novel disclosure risks, in both type and scale. Objectives: As part of a series on ML disclosure risk in TREs, this article is intended to introduce TRE managers to the conceptual problems and work being done to address them. Methods: We demonstrate how ML models present a qualitatively different type of disclosure risk, compared to traditional statistical outputs. These arise from both the nature and the scale of ML modelling. Results: We show that there are a large number of unresolved issues, although there is progress in many areas. We show where areas of uncertainty remain, as well as remedial responses available to TREs. Conclusions: At this stage, disclosure checking of ML models is very much a specialist activity. However, TRE managers need a basic awareness of the potential risk in ML models to enable them to make sensible decisions on using TREs for ML model development.

A Hybrid Architecture (CO-CONNECT) to Facilitate Rapid Discovery and Access to Data Across the United Kingdom in Response to the COVID-19 Pandemic: Development Study.

BACKGROUND: COVID-19 data have been generated across the United Kingdom as a by-product of clinical care and public health provision, as well as numerous bespoke and repurposed research endeavors. Analysis of these data has underpinned the United Kingdom's response to the pandemic, and informed public health policies and clinical guidelines. However, these data are held by different organizations, and this fragmented landscape has presented challenges for public health agencies and researchers as they struggle to find relevant data to access and interrogate the data they need to inform the pandemic response at pace. OBJECTIVE: We aimed to transform UK COVID-19 diagnostic data sets to be findable, accessible, interoperable, and reusable (FAIR). METHODS: A federated infrastructure model (COVID - Curated and Open Analysis and Research Platform [CO-CONNECT]) was rapidly built to enable the automated and reproducible mapping of health data partners' pseudonymized data to the Observational Medical Outcomes Partnership Common Data Model without the need for any data to leave the data controllers' secure environments, and to support federated cohort discovery queries and meta-analysis. RESULTS: A total of 56 data sets from 19 organizations are being connected to the federated network. The data include research cohorts and COVID-19 data collected through routine health care provision linked to longitudinal health care records and demographics. The infrastructure is live, supporting aggregate-level querying of data across the United Kingdom. CONCLUSIONS: CO-CONNECT was developed by a multidisciplinary team. It enables rapid COVID-19 data discovery and instantaneous meta-analysis across data sources, and it is researching streamlined data extraction for use in a Trusted Research Environment for research and public health analysis. CO-CONNECT has the potential to make UK health data more interconnected and better able to answer national-level research questions while maintaining patient confidentiality and local governance procedures.

Machine learning for classification of hypertension subtypes using multi-omics: A multi-centre, retrospective, data-driven study.

BACKGROUND: Arterial hypertension is a major cardiovascular risk factor. Identification of secondary hypertension in its various forms is key to preventing and targeting treatment of cardiovascular complications. Simplified diagnostic tests are urgently required to distinguish primary and secondary hypertension to address the current underdiagnosis of the latter. METHODS: This study uses Machine Learning (ML) to classify subtypes of endocrine hypertension (EHT) in a large cohort of hypertensive patients using multidimensional omics analysis of plasma and urine samples. We measured 409 multi-omics (MOmics) features including plasma miRNAs (PmiRNA: 173), plasma catechol O-methylated metabolites (PMetas: 4), plasma steroids (PSteroids: 16), urinary steroid metabolites (USteroids: 27), and plasma small metabolites (PSmallMB: 189) in primary hypertension (PHT) patients, EHT patients with either primary aldosteronism (PA), pheochromocytoma/functional paraganglioma (PPGL) or Cushing syndrome (CS) and normotensive volunteers (NV). Biomarker discovery involved selection of disease combination, outlier handling, feature reduction, 8 ML classifiers, class balancing and consideration of different age- and sex-based scenarios. Classifications were evaluated using balanced accuracy, sensitivity, specificity, AUC, F1, and Kappa score. FINDINGS: Complete clinical and biological datasets were generated from 307 subjects (PA=113, PPGL=88, CS=41 and PHT=112). The random forest classifier provided ∼92% balanced accuracy (∼11% improvement on the best mono-omics classifier), with 96% specificity and 0.95 AUC to distinguish one of the four conditions in multi-class ALL-ALL comparisons (PPGL vs PA vs CS vs PHT) on an unseen test set, using 57 MOmics features. For discrimination of EHT (PA + PPGL + CS) vs PHT, the simple logistic classifier achieved 0.96 AUC with 90% sensitivity, and ∼86% specificity, using 37 MOmics features. One PmiRNA (hsa-miR-15a-5p) and two PSmallMB (C9 and PC ae C38:1) features were found to be most discriminating for all disease combinations. Overall, the MOmics-based classifiers were able to provide better classification performance in comparison to mono-omics classifiers. INTERPRETATION: We have developed a ML pipeline to distinguish different EHT subtypes from PHT using multi-omics data. This innovative approach to stratification is an advancement towards the development of a diagnostic tool for EHT patients, significantly increasing testing throughput and accelerating administration of appropriate treatment. FUNDING: European Union's Horizon 2020 Research and Innovation Programme under Grant Agreement No. 633983, Clinical Research Priority Program of the University of Zurich for the CRPP HYRENE (to Z.E. and F.B.), and Deutsche Forschungsgemeinschaft (CRC/Transregio 205/1).

A National Network of Safe Havens: Scottish Perspective.

For over a decade, Scotland has implemented and operationalized a system of Safe Havens, which provides secure analytics platforms for researchers to access linked, deidentified electronic health records (EHRs) while managing the risk of unauthorized reidentification. In this paper, a perspective is provided on the state-of-the-art Scottish Safe Haven network, including its evolution, to define the key activities required to scale the Scottish Safe Haven network's capability to facilitate research and health care improvement initiatives. A set of processes related to EHR data and their delivery in Scotland have been discussed. An interview with each Safe Haven was conducted to understand their services in detail, as well as their commonalities. The results show how Safe Havens in Scotland have protected privacy while facilitating the reuse of the EHR data. This study provides a common definition of a Safe Haven and promotes a consistent understanding among the Scottish Safe Haven network and the clinical and academic research community. We conclude by identifying areas where efficiencies across the network can be made to meet the needs of population-level studies at scale.

Creation of a universal experimental protocol for the investigation of transfer and persistence of trace evidence: Part 2 - Implementation and preliminary data.

This is the second paper on the development and implementation of a universal experimental protocol for transfer and persistence of trace evidence. Here, we present the results of five individual researchers who implemented the universal experimental protocol for the first time. Over 2500 images were collected, computationally analysed and statistically compared. The results were shown to be reliable and consistent under all conditions tested and were used to model the rate of loss of transferred particles over a 7-day timescale. The protocol was additionally extended to include a test of camera settings. The protocol was found to be useable and robust in this preliminary trial paving the way for it to be deployed more widely.

Creation of a universal experimental protocol for the investigation of transfer and persistence of trace evidence: Part 1 - From design to implementation for particulate evidence.

Understanding the transfer and persistence of different types of trace evidence between different donor and receiving surfaces under specific conditions, circumstances and alleged competing defence and prosecution hypotheses is a significant need. Acquiring such a knowledge base enables hypothesis testing to be undertaken more readily and with greater confidence. A longstanding goal has been to develop a unified approach to transfer and persistence studies which are fit for purpose but also scalable. Here we propose a low cost, universal experimental protocol using a recognised and well researched proxy material for the development and aggregation of ground truth transfer and persistence data at scale. We also propose and provide the tools to enable the creation of an open source and open access data repository of experimental data to act as a resource for practitioners and researchers in addressing transfer and persistence questions.

EMSY expression affects multiple components of the skin barrier with relevance to atopic dermatitis.

BACKGROUND: Atopic dermatitis (AD) is a common, complex, and highly heritable inflammatory skin disease. Genome-wide association studies offer opportunities to identify molecular targets for drug development. A risk locus on chromosome 11q13.5 lies between 2 candidate genes, EMSY and LRRC32 (leucine-rich repeat-containing 32) but the functional mechanisms affecting risk of AD remain unclear. OBJECTIVES: We sought to apply a combination of genomic and molecular analytic techniques to investigate which genes are responsible for genetic risk at this locus and to define mechanisms contributing to atopic skin disease. METHODS: We used interrogation of available genomic and chromosome conformation data in keratinocytes, small interfering RNA (siRNA)-mediated knockdown in skin organotypic culture and functional assessment of barrier parameters, mass spectrometric global proteomic analysis and quantitative lipid analysis, electron microscopy of organotypic skin, and immunohistochemistry of human skin samples. RESULTS: Genomic data indicate active promoters in the genome-wide association study locus and upstream of EMSY; EMSY, LRRC32, and intergenic variants all appear to be within a single topologically associating domain. siRNA-knockdown of EMSY in organotypic culture leads to enhanced development of barrier function, reflecting increased expression of structural and functional proteins, including filaggrin and filaggrin-2, as well as long-chain ceramides. Conversely, overexpression of EMSY in keratinocytes leads to a reduction in markers of barrier formation. Skin biopsy samples from patients with AD show greater EMSY staining in the nucleus, which is consistent with an increased functional effect of this transcriptional control protein. CONCLUSION: Our findings demonstrate an important role for EMSY in transcriptional regulation and skin barrier formation, supporting EMSY inhibition as a therapeutic approach.

Optimization of ClpXP activity and protein synthesis in an E. coli extract-based cell-free expression system.

Protein degradation is a fundamental process in all living cells and is essential to remove both damaged proteins and intact proteins that are no longer needed by the cell. We are interested in creating synthetic genetic circuits that function in a cell-free expression system. This will require not only an efficient protein expression platform but also a robust protein degradation system in cell extract. Therefore, we purified and tested the activity of E. coli ClpXP protease in cell-free transcription-translation (TX-TL) systems that used E. coli S30 cell extract. Surprisingly, our studies showed that purified ClpXP added to the TX-TL system has very low proteolytic activity. The low activity of ClpXP was correlated with the rapid consumption of adenosine triphosphate (ATP) in cell extract. We improved the activity of ClpXP in cell extract by adding exogenous ATP and an energy regeneration system. We then established conditions for both protein synthesis, and protein degradation by ClpXP to occur simultaneously in the TX-TL systems. The optimized conditions for ClpXP activity will be useful for creating tunable synthetic genetic circuits and in vitro synthetic biology.

In Situ Reconstitution of the Adenosine A2A Receptor in Spontaneously Formed Synthetic Liposomes.

Cell transmembrane receptors play a key role in the detection of environmental stimuli and control of intracellular communication. G protein-coupled receptors constitute the largest transmembrane protein family involved in cell signaling. However, current methods for their functional reconstitution in biomimetic membranes remain both challenging and limited in scope. Herein, we describe the spontaneous reconstitution of adenosine A2A receptor (A2AR) during the de novo formation of synthetic liposomes via native chemical ligation. The approach takes advantage of a nonenzymatic and chemoselective method to rapidly generate A2AR embedded phospholiposomes from receptor solubilized in n-dodecyl-ß-d-maltoside analogs. In situ lipid synthesis for protein reconstitution technology proceeds in the absence of dialysis and/or detergent absorbents, and A2AR assimilation into synthetic liposomes can be visualized by microscopy and probed by radio-ligand binding.

Improved annotation with de novo transcriptome assembly in four social amoeba species.

BACKGROUND: Annotation of gene models and transcripts is a fundamental step in genome sequencing projects. Often this is performed with automated prediction pipelines, which can miss complex and atypical genes or transcripts. RNA sequencing (RNA-seq) data can aid the annotation with empirical data. Here we present de novo transcriptome assemblies generated from RNA-seq data in four Dictyostelid species: D. discoideum, P. pallidum, D. fasciculatum and D. lacteum. The assemblies were incorporated with existing gene models to determine corrections and improvement on a whole-genome scale. This is the first time this has been performed in these eukaryotic species. RESULTS: An initial de novo transcriptome assembly was generated by Trinity for each species and then refined with Program to Assemble Spliced Alignments (PASA). The completeness and quality were assessed with the Benchmarking Universal Single-Copy Orthologs (BUSCO) and Transrate tools at each stage of the assemblies. The final datasets of 11,315-12,849 transcripts contained 5,610-7,712 updates and corrections to >50% of existing gene models including changes to hundreds or thousands of protein products. Putative novel genes are also identified and alternative splice isoforms were observed for the first time in P. pallidum, D. lacteum and D. fasciculatum. CONCLUSIONS: In taking a whole transcriptome approach to genome annotation with empirical data we have been able to enrich the annotations of four existing genome sequencing projects. In doing so we have identified updates to the majority of the gene annotations across all four species under study and found putative novel genes and transcripts which could be worthy for follow-up. The new transcriptome data we present here will be a valuable resource for genome curators in the Dictyostelia and we propose this effective methodology for use in other genome annotation projects.

Adenylate cyclase A acting on PKA mediates induction of stalk formation by cyclic diguanylate at the Dictyostelium organizer.

Coordination of cell movement with cell differentiation is a major feat of embryonic development. The Dictyostelium stalk always forms at the organizing tip, by a mechanism that is not understood. We previously reported that cyclic diguanylate (c-di-GMP), synthesized by diguanylate cyclase A (DgcA), induces stalk formation. Here we used transcriptional profiling of dgca- structures to identify target genes for c-di-GMP, and used these genes to investigate the c-di-GMP signal transduction pathway. We found that knockdown of cAMP-dependent protein kinase (PKA) activity in prestalk cells reduced stalk gene induction by c-di-GMP, whereas PKA activation bypassed the c-di-GMP requirement for stalk gene expression. c-di-GMP caused a persistent increase in cAMP, which still occurred in mutants lacking the adenylate cyclases ACG or ACR, or the cAMP phosphodiesterase RegA. However, both inhibition of adenylate cyclase A (ACA) with SQ22536 and incubation of a temperature-sensitive ACA mutant at the restrictive temperature prevented c-di-GMP-induced cAMP synthesis as well as c-di-GMP-induced stalk gene transcription. ACA produces the cAMP pulses that coordinate Dictyostelium morphogenetic cell movement and is highly expressed at the organizing tip. The stalk-less dgca- mutant regained its stalk by expression of a light-activated adenylate cyclase from the ACA promoter and exposure to light, indicating that cAMP is also the intermediate for c-di-GMP in vivo. Our data show that the more widely expressed DgcA activates tip-expressed ACA, which then acts on PKA to induce stalk genes. These results explain why stalk formation in Dictyostelia always initiates at the site of the morphogenetic organizer.

A mechanistic target of rapamycin complex 1/2 (mTORC1)/V-Akt murine thymoma viral oncogene homolog 1 (AKT1)/cathepsin H axis controls filaggrin expression and processing in skin, a novel mechanism for skin barrier disruption in patients with atopic dermatitis.

BACKGROUND: Filaggrin, which is encoded by the filaggrin gene (FLG), is an important component of the skin's barrier to the external environment, and genetic defects in FLG strongly associate with atopic dermatitis (AD). However, not all patients with AD have FLG mutations. OBJECTIVE: We hypothesized that these patients might possess other defects in filaggrin expression and processing contributing to barrier disruption and AD, and therefore we present novel therapeutic targets for this disease. RESULTS: We describe the relationship between the mechanistic target of rapamycin complex 1/2 protein subunit regulatory associated protein of the MTOR complex 1 (RAPTOR), the serine/threonine kinase V-Akt murine thymoma viral oncogene homolog 1 (AKT1), and the protease cathepsin H (CTSH), for which we establish a role in filaggrin expression and processing. Increased RAPTOR levels correlated with decreased filaggrin expression in patients with AD. In keratinocyte cell cultures RAPTOR upregulation or AKT1 short hairpin RNA knockdown reduced expression of the protease CTSH. Skin of CTSH-deficient mice and CTSH short hairpin RNA knockdown keratinocytes showed reduced filaggrin processing, and the mouse had both impaired skin barrier function and a mild proinflammatory phenotype. CONCLUSION: Our findings highlight a novel and potentially treatable signaling axis controlling filaggrin expression and processing that is defective in patients with AD.

A Proteomic Approach to Analyze the Aspirin-mediated Lysine Acetylome.

Aspirin, or acetylsalicylic acid is widely used to control pain, inflammation and fever. Important to this function is its ability to irreversibly acetylate cyclooxygenases at active site serines. Aspirin has the potential to acetylate other amino acid side-chains, leading to the possibility that aspirin-mediated lysine acetylation could explain some of its as-yet unexplained drug actions or side-effects. Using isotopically labeled aspirin-d3, in combination with acetylated lysine purification and LC-MS/MS, we identified over 12000 sites of lysine acetylation from cultured human cells. Although aspirin amplifies endogenous acetylation signals at the majority of detectable endogenous sites, cells tolerate aspirin mediated acetylation very well unless cellular deacetylases are inhibited. Although most endogenous acetylations are amplified by orders of magnitude, lysine acetylation site occupancies remain very low even after high doses of aspirin. This work shows that while aspirin has enormous potential to alter protein function, in the majority of cases aspirin-mediated acetylations do not accumulate to levels likely to elicit biological effects. These findings are consistent with an emerging model for cellular acetylation whereby stoichiometry correlates with biological relevance, and deacetylases act to minimize the biological consequences of nonspecific chemical acetylations.

AlmostSignificant: simplifying quality control of high-throughput sequencing data.

MOTIVATION: The current generation of DNA sequencing technologies produce a large amount of data quickly. All of these data need to pass some form of quality control (QC) processing and checking before they can be used for any analysis. The large number of samples that are run through Illumina sequencing machines makes the process of QC an onerous and time-consuming task that requires multiple pieces of information from several sources. RESULTS: AlmostSignificant is an open-source platform for aggregating multiple sources of quality metrics as well as run and sample meta-data associated with DNA sequencing runs from Illumina sequencing machines. AlmostSignificant is a graphical platform to streamline the QC of DNA sequencing data, to store these data for future reference together with extra meta-data associated with the sequencing runs not typically retained. This simplifies the challenge of monitoring the volume of data produced by Illumina sequencers. AlmostSignificant has been used to track the quality of over 80 sequencing runs covering over 2500 samples produced over the last three years. AVAILABILITY AND IMPLEMENTATION: The code and documentation for AlmostSignificant is freely available at https://github.com/bartongroup/AlmostSignificant CONTACTS: c.cole@dundee.ac.uk or g.j.barton@dundee.ac.ukSupplementary information: Supplementary data are available at Bioinformatics online.

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.