Search

Background: Obesity rates have nearly tripled in the past 50 years, and by 2030 more than 1 billion individuals worldwide are projected to be obese. This creates a significant economic strain due to the associated non-communicable diseases. The root cause is an energy expenditure imbalance, owing to an interplay of lifestyle, environmental, and genetic factors. Obesity has a polygenic genetic architecture; however, single genetic variants with large effect size are etiological in a minority of cases. These variants allowed the discovery of novel genes and biology relevant to weight regulation and ultimately led to the development of novel specific treatments., Methods: We used a case-control approach to determine metabolic differences between individuals homozygous for a loss-of-function genetic variant in the small integral membrane protein 1 (SMIM1) and the general population, leveraging data from five cohorts. Metabolic characterization of SMIM1 -/- individuals was performed using plasma biochemistry, calorimetric chamber, and DXA scan., Findings: We found that individuals homozygous for a loss-of-function genetic variant in SMIM1 gene, underlying the blood group Vel, display excess body weight, dyslipidemia, altered leptin to adiponectin ratio, increased liver enzymes, and lower thyroid hormone levels. This was accompanied by a reduction in resting energy expenditure., Conclusion: This research identified a novel genetic predisposition to being overweight or obese. It highlights the need to investigate the genetic causes of obesity to select the most appropriate treatment given the large cost disparity between them., Funding: This work was funded by the National Institute of Health Research, British Heart Foundation, and NHS Blood and Transplant., Competing Interests: Declaration of interests J.S. is the deputy CEO and 50% owner of BLUsang AB. He holds patents on Vel genotyping (inventors: Jill Storry, Magnus Jöud, Björn Nilsson, and Martin L. Olsson). J.S. has received speaker fees, royalties, and honoraria from the following companies: Grifols Diagnostic Solutions, QuidelOrtho Inc., and Biorad Laboratories. J.S. receives an honorarium for Section Editor work, Vox Sanguinis from John Wiley & Sons Ltd. J.S. is Vice President of the International Society of Blood Transfusion and married to Professor M.L.O. M.L.O. is CEO and 50% owner of BLUsang AB. M.L.O. holds patents on Vel genotyping (inventors: Jill Storry, Magnus Jöud, Björn Nilsson, and Martin L. Olsson). M.L.O. received speaker fees, royalties, and honoraria from the following companies: Grifols Diagnostic Solutions, QuidelOrtho Inc., and Biorad Laboratories. M.L.O. is married to Adjunct Professor J.S. W.N.E. is chair of the International Council for Standardization in Haematology. W.N.E. works as advisor for Scorpio Labs and is on the editorial board of the Journal of Clinical Pathology. W.H.O. is chair of the Blood Transfusion Genomics Consortium. W.H.O. is in receipt of an educational/research grant from Thermo Fisher Scientific. N.G. offers scientific consulting services to Thermo Fisher Scientific., (Copyright © 2024 The Author(s). Published by Elsevier Inc. All rights reserved.)

Genetic studies of platelet reactivity (PR) phenotypes may identify novel antiplatelet drug targets. However, such studies have been limited by small sample sizes (n < 5000) because of the complexity of measuring PR. We trained a model to predict PR from complete blood count (CBC) scattergrams. A genome-wide association study of this phenotype in 29 806 blood donors identified 21 distinct associations implicating 20 genes, of which 6 have been identified previously. The effect size estimates were significantly correlated with estimates from a study of flow cytometry-measured PR and a study of a phenotype of in vitro thrombus formation. A genetic score of PR built from the 21 variants was associated with the incidence rates of myocardial infarction and pulmonary embolism. Mendelian randomization analyses showed that PR was causally associated with the risks of coronary artery disease, stroke, and venous thromboembolism. Our approach provides a blueprint for using phenotype imputation to study the determinants of hard-to-measure but biologically important hematological traits., (© 2023 by The American Society of Hematology. Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0), permitting only noncommercial, nonderivative use with attribution. All other rights reserved.)

Blood cells contain functionally important intracellular structures, such as granules, critical to immunity and thrombosis. Quantitative variation in these structures has not been subjected previously to large-scale genetic analysis. We perform genome-wide association studies of 63 flow-cytometry derived cellular phenotypes-including cell-type specific measures of granularity, nucleic acid content and reactivity-in 41,515 participants in the INTERVAL study. We identify 2172 distinct variant-trait associations, including associations near genes coding for proteins in organelles implicated in inflammatory and thrombotic diseases. By integrating with epigenetic data we show that many intracellular structures are likely to be determined in immature precursor cells. By integrating with proteomic data we identify the transcription factor FOG2 as an early regulator of platelet formation and α-granularity. Finally, we show that colocalisation of our associations with disease risk signals can suggest aetiological cell-types-variants in IL2RA and ITGA4 respectively mirror the known effects of daclizumab in multiple sclerosis and vedolizumab in inflammatory bowel disease., (© 2023. Springer Nature Limited.)

There is an increasing body of work exploring the integration of random projection into algorithms for numerical linear algebra. The primary motivation is to reduce the overall computational cost of processing large datasets. A suitably chosen random projection can be used to embed the original dataset in a lower-dimensional space such that key properties of the original dataset are retained. These algorithms are often referred to as sketching algorithms, as the projected dataset can be used as a compressed representation of the full dataset. We show that random matrix theory, in particular the Tracy-Widom law, is useful for describing the operating characteristics of sketching algorithms in the tall-data regime when the sample size n is much greater than the number of variables d . Asymptotic large sample results are of particular interest as this is the regime where sketching is most useful for data compression. In particular, we develop asymptotic approximations for the success rate in generating random subspace embeddings and the convergence probability of iterative sketching algorithms. We test a number of sketching algorithms on real large high-dimensional datasets and find that the asymptotic expressions give accurate predictions of the empirical performance., Supplementary Information: The online version contains supplementary material available at 10.1007/s11222-022-10148-5., Competing Interests: Conflict of interestThe authors have no competing interests to declare that are relevant to the content of this article., (© The Author(s) 2023.)

The interindividual variation in the functional response of platelets to activation by agonists is heritable. Genome-wide association studies (GWASs) of quantitative measures of platelet function have identified fewer than 20 distinctly associated variants, some with unknown mechanisms. Here, we report GWASs of pathway-specific functional responses to agonism by adenosine 5'-diphosphate, a glycoprotein VI-specific collagen mimetic, and thrombin receptor-agonist peptides, each specific to 1 of the G protein-coupled receptors PAR-1 and PAR-4, in subsets of 1562 individuals. We identified an association (P = 2.75 × 10-40) between a common intronic variant, rs10886430, in the G protein-coupled receptor kinase 5 gene (GRK5) and the sensitivity of platelets to activate through PAR-1. The variant resides in a megakaryocyte-specific enhancer that is bound by the transcription factors GATA1 and MEIS1. The minor allele (G) is associated with fewer GRK5 transcripts in platelets and the greater sensitivity of platelets to activate through PAR-1. We show that thrombin-mediated activation of human platelets causes binding of GRK5 to PAR-1 and that deletion of the mouse homolog Grk5 enhances thrombin-induced platelet activation sensitivity and increases platelet accumulation at the site of vascular injury. This corroborates evidence that the human G allele of rs10886430 is associated with a greater risk for cardiovascular disease. In summary, by combining the results of pathway-specific GWASs and expression quantitative trait locus studies in humans with the results from platelet function studies in Grk5-/- mice, we obtain evidence that GRK5 regulates the human platelet response to thrombin via the PAR-1 pathway., (© 2022 by The American Society of Hematology. Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0), permitting only noncommercial, nonderivative use with attribution. All other rights reserved.)

Low platelet count, or thrombocytopenia, is a common haematological abnormality, with a wide differential diagnosis, which may represent a clinically significant underlying pathology. Macrothrombocytopenia, the presence of large platelets in combination with thrombocytopenia, can be acquired or hereditary and indicative of a complex disorder. In this review, we discuss the interpretation of platelet count and volume measured by automated haematology analysers and highlight some important technical considerations relevant to the analysis of blood samples with macrothrombocytopenia. We review how large cohorts, such as the UK Biobank and INTERVAL studies, have enabled an accurate description of the distribution and co-variation of platelet parameters in adult populations. We discuss how genome-wide association studies have identified hundreds of genetic associations with platelet count and mean platelet volume, which in aggregate can explain large fractions of phenotypic variance, consistent with a complex genetic architecture and polygenic inheritance. Finally, we describe the large genetic diagnostic and discovery programmes, which, simultaneously to genome-wide association studies, have expanded the repertoire of genes and variants associated with extreme platelet phenotypes. These have advanced our understanding of the pathogenesis of hereditary macrothrombocytopenia and support a future clinical diagnostic strategy that utilises genotype alongside clinical and laboratory phenotype data., (© 2021 The Authors. British Journal of Haematology published by British Society for Haematology and John Wiley & Sons Ltd.)

Sketching is a probabilistic data compression technique that has been largely developed by the computer science community. Numerical operations on big datasets can be intolerably slow; sketching algorithms address this issue by generating a smaller surrogate dataset. Typically, inference proceeds on the compressed dataset. Sketching algorithms generally use random projections to compress the original dataset, and this stochastic generation process makes them amenable to statistical analysis. We argue that the sketched data can be modelled as a random sample, thus placing this family of data compression methods firmly within an inferential framework. In particular, we focus on the Gaussian, Hadamard and Clarkson-Woodruff sketches and their use in single-pass sketching algorithms for linear regression with huge samples. We explore the statistical properties of sketched regression algorithms and derive new distributional results for a large class of sketching estimators. A key result is a conditional central limit theorem for data-oblivious sketches. An important finding is that the best choice of sketching algorithm in terms of mean squared error is related to the signal-to-noise ratio in the source dataset. Finally, we demonstrate the theory and the limits of its applicability on two datasets.

Iron is essential for many biological functions and iron deficiency and overload have major health implications. We performed a meta-analysis of three genome-wide association studies from Iceland, the UK and Denmark of blood levels of ferritin (N = 246,139), total iron binding capacity (N = 135,430), iron (N = 163,511) and transferrin saturation (N = 131,471). We found 62 independent sequence variants associating with iron homeostasis parameters at 56 loci, including 46 novel loci. Variants at DUOX2, F5, SLC11A2 and TMPRSS6 associate with iron deficiency anemia, while variants at TF, HFE, TFR2 and TMPRSS6 associate with iron overload. A HBS1L-MYB intergenic region variant associates both with increased risk of iron overload and reduced risk of iron deficiency anemia. The DUOX2 missense variant is present in 14% of the population, associates with all iron homeostasis biomarkers, and increases the risk of iron deficiency anemia by 29%. The associations implicate proteins contributing to the main physiological processes involved in iron homeostasis: iron sensing and storage, inflammation, absorption of iron from the gut, iron recycling, erythropoiesis and bleeding/menstruation.

Gray platelet syndrome (GPS) is a rare recessive disorder caused by biallelic variants in NBEAL2 and characterized by bleeding symptoms, the absence of platelet α-granules, splenomegaly, and bone marrow (BM) fibrosis. Due to the rarity of GPS, it has been difficult to fully understand the pathogenic processes that lead to these clinical sequelae. To discern the spectrum of pathologic features, we performed a detailed clinical genotypic and phenotypic study of 47 patients with GPS and identified 32 new etiologic variants in NBEAL2. The GPS patient cohort exhibited known phenotypes, including macrothrombocytopenia, BM fibrosis, megakaryocyte emperipolesis of neutrophils, splenomegaly, and elevated serum vitamin B12 levels. Novel clinical phenotypes were also observed, including reduced leukocyte counts and increased presence of autoimmune disease and positive autoantibodies. There were widespread differences in the transcriptome and proteome of GPS platelets, neutrophils, monocytes, and CD4 lymphocytes. Proteins less abundant in these cells were enriched for constituents of granules, supporting a role for Nbeal2 in the function of these organelles across a wide range of blood cells. Proteomic analysis of GPS plasma showed increased levels of proteins associated with inflammation and immune response. One-quarter of plasma proteins increased in GPS are known to be synthesized outside of hematopoietic cells, predominantly in the liver. In summary, our data show that, in addition to the well-described platelet defects in GPS, there are immune defects. The abnormal immune cells may be the drivers of systemic abnormalities such as autoimmune disease., (© 2020 by The American Society of Hematology.)

Blood cells play essential roles in human health, underpinning physiological processes such as immunity, oxygen transport, and clotting, which when perturbed cause a significant global health burden. Here we integrate data from UK Biobank and a large-scale international collaborative effort, including data for 563,085 European ancestry participants, and discover 5,106 new genetic variants independently associated with 29 blood cell phenotypes covering a range of variation impacting hematopoiesis. We holistically characterize the genetic architecture of hematopoiesis, assess the relevance of the omnigenic model to blood cell phenotypes, delineate relevant hematopoietic cell states influenced by regulatory genetic variants and gene networks, identify novel splice-altering variants mediating the associations, and assess the polygenic prediction potential for blood traits and clinical disorders at the interface of complex and Mendelian genetics. These results show the power of large-scale blood cell trait GWAS to interrogate clinically meaningful variants across a wide allelic spectrum of human variation., Competing Interests: Declaration of Interests Adam Butterworth has received grants (outside of this work) from AstraZeneca, Biogen, BioMarin, Bioverativ, Merck, Novartis, and Sanofi; James Floyd has consulted for Shionogi; Qi Guo is a full-time employee of BenevolentAI; Joanna Howson is a full-time employee of Novo Nordisk. Parsa Akbari is a full-time employee of Regeneron Pharmaceuticals., (Copyright © 2020 The Authors. Published by Elsevier Inc. All rights reserved.)

Most loci identified by GWASs have been found in populations of European ancestry (EUR). In trans-ethnic meta-analyses for 15 hematological traits in 746,667 participants, including 184,535 non-EUR individuals, we identified 5,552 trait-variant associations at p < 5 × 10 -9 , including 71 novel associations not found in EUR populations. We also identified 28 additional novel variants in ancestry-specific, non-EUR meta-analyses, including an IL7 missense variant in South Asians associated with lymphocyte count in vivo and IL-7 secretion levels in vitro. Fine-mapping prioritized variants annotated as functional and generated 95% credible sets that were 30% smaller when using the trans-ethnic as opposed to the EUR-only results. We explored the clinical significance and predictive value of trans-ethnic variants in multiple populations and compared genetic architecture and the effect of natural selection on these blood phenotypes between populations. Altogether, our results for hematological traits highlight the value of a more global representation of populations in genetic studies., Competing Interests: Declaration of Interests Competing financial interests are declared in Table S1F., (Copyright © 2020 Elsevier Inc. All rights reserved.)

Each year, blood transfusions save millions of lives. However, under current blood-matching practices, sensitization to non-self-antigens is an unavoidable adverse side effect of transfusion. We describe a universal donor typing platform that could be adopted by blood services worldwide to facilitate a universal extended blood-matching policy and reduce sensitization rates. This DNA-based test is capable of simultaneously typing most clinically relevant red blood cell (RBC), human platelet (HPA), and human leukocyte (HLA) antigens. Validation was performed, using samples from 7927 European, 27 South Asian, 21 East Asian, and 9 African blood donors enrolled in 2 national biobanks. We illustrated the usefulness of the platform by analyzing antibody data from patients sensitized with multiple RBC alloantibodies. Genotyping results demonstrated concordance of 99.91%, 99.97%, and 99.03% with RBC, HPA, and HLA clinically validated typing results in 89 371, 3016, and 9289 comparisons, respectively. Genotyping increased the total number of antigen typing results available from 110 980 to >1 200 000. Dense donor typing allowed identification of 2 to 6 times more compatible donors to serve 3146 patients with multiple RBC alloantibodies, providing at least 1 match for 176 individuals for whom previously no blood could be found among the same donors. This genotyping technology is already being used to type thousands of donors taking part in national genotyping studies. Extraction of dense antigen-typing data from these cohorts provides blood supply organizations with the opportunity to implement a policy of genomics-based precision matching of blood.

Most patients with rare diseases do not receive a molecular diagnosis and the aetiological variants and causative genes for more than half such disorders remain to be discovered 1 . Here we used whole-genome sequencing (WGS) in a national health system to streamline diagnosis and to discover unknown aetiological variants in the coding and non-coding regions of the genome. We generated WGS data for 13,037 participants, of whom 9,802 had a rare disease, and provided a genetic diagnosis to 1,138 of the 7,065 extensively phenotyped participants. We identified 95 Mendelian associations between genes and rare diseases, of which 11 have been discovered since 2015 and at least 79 are confirmed to be aetiological. By generating WGS data of UK Biobank participants 2 , we found that rare alleles can explain the presence of some individuals in the tails of a quantitative trait for red blood cells. Finally, we identified four novel non-coding variants that cause disease through the disruption of transcription of ARPC1B, GATA1, LRBA and MPL. Our study demonstrates a synergy by using WGS for diagnosis and aetiological discovery in routine healthcare.

Mutations in NBEAL2 , the gene encoding the scaffolding protein Nbeal2, are causal of gray platelet syndrome (GPS), a rare recessive bleeding disorder characterized by platelets lacking α-granules and progressive marrow fibrosis. We present here the interactome of Nbeal2 with additional validation by reverse immunoprecipitation of Dock7, Sec16a, and Vac14 as interactors of Nbeal2. We show that GPS-causing mutations in its BEACH domain have profound and possible effects on the interaction with Dock7 and Vac14, respectively. Proximity ligation assays show that these 2 proteins are physically proximal to Nbeal2 in human megakaryocytes. In addition, we demonstrate that Nbeal2 is primarily localized in the cytoplasm and Dock7 on the membrane of or in α-granules. Interestingly, platelets from GPS cases and Nbeal2 -/- mice are almost devoid of Dock7, resulting in a profound dysregulation of its signaling pathway, leading to defective actin polymerization, platelet activation, and shape change. This study shows for the first time proteins interacting with Nbeal2 and points to the dysregulation of the canonical signaling pathway of Dock7 as a possible cause of the aberrant formation of platelets in GPS cases and Nbeal2- deficient mice., (© 2018 by The American Society of Hematology.)

Linking non-coding genetic variants associated with the risk of diseases or disease-relevant traits to target genes is a crucial step to realize GWAS potential in the introduction of precision medicine. Here we set out to determine the mechanisms underpinning variant association with platelet quantitative traits using cell type-matched epigenomic data and promoter long-range interactions. We identify potential regulatory functions for 423 of 565 (75%) non-coding variants associated with platelet traits and we demonstrate, through ex vivo and proof of principle genome editing validation, that variants in super enhancers play an important role in controlling archetypical platelet functions.

Genome-wide association studies have identified a genetic variant at 3p14.3 (SNP rs1354034) that strongly associates with platelet number and mean platelet volume in humans. While originally proposed to be intronic, analysis of mRNA expression in primary human hematopoietic subpopulations reveals that this SNP is located directly upstream of the predominantly expressed ARHGEF3 isoform in megakaryocytes (MK). We found that ARHGEF3, which encodes a Rho guanine exchange factor, is dramatically upregulated during both human and murine MK maturation. We show that the SNP (rs1354034) is located in a DNase I hypersensitive region in human MKs and is an expression quantitative locus (eQTL) associated with ARHGEF3 expression level in human platelets, suggesting that it may be the causal SNP that accounts for the variations observed in human platelet traits and ARHGEF3 expression. In vitro human platelet activation assays revealed that rs1354034 is highly correlated with human platelet activation by ADP. In order to test whether ARHGEF3 plays a role in MK development and/or platelet function, we developed an Arhgef3 KO/LacZ reporter mouse model. Reflecting changes in gene expression, LacZ expression increases during MK maturation in these mice. Although Arhgef3 KO mice have significantly larger platelets, loss of Arhgef3 does not affect baseline MK or platelets nor does it affect platelet function or platelet recovery in response to antibody-mediated platelet depletion compared to littermate controls. In summary, our data suggest that modulation of ARHGEF3 gene expression in humans with a promoter-localized SNP plays a role in human MKs and human platelet function-a finding resulting from the biological follow-up of human genetic studies. Arhgef3 KO mice partially recapitulate the human phenotype.

Platelets are anuclear cells that are essential for blood clotting. They are produced by large polyploid precursor cells called megakaryocytes. Previous genome-wide association studies in nearly 70,000 individuals indicated that single nucleotide variants (SNVs) in the gene encoding the actin cytoskeletal regulator tropomyosin 4 (TPM4) exert an effect on the count and volume of platelets. Platelet number and volume are independent risk factors for heart attack and stroke. Here, we have identified 2 unrelated families in the BRIDGE Bleeding and Platelet Disorders (BPD) collection who carry a TPM4 variant that causes truncation of the TPM4 protein and segregates with macrothrombocytopenia, a disorder characterized by low platelet count. N-Ethyl-N-nitrosourea-induced (ENU-induced) missense mutations in Tpm4 or targeted inactivation of the Tpm4 locus led to gene dosage-dependent macrothrombocytopenia in mice. All other blood cell counts in Tpm4-deficient mice were normal. Insufficient TPM4 expression in human and mouse megakaryocytes resulted in a defect in the terminal stages of platelet production and had a mild effect on platelet function. Together, our findings demonstrate a nonredundant role for TPM4 in platelet biogenesis in humans and mice and reveal that truncating variants in TPM4 cause a previously undescribed dominant Mendelian platelet disorder.

The von Willebrand receptor complex, which is composed of the glycoproteins Ibα, Ibβ, GPV, and GPIX, plays an essential role in the earliest steps in hemostasis. During the last 4 decades, it has become apparent that loss of function of any 1 of 3 of the genes encoding these glycoproteins (namely, GP1BA, GP1BB, and GP9) leads to autosomal recessive macrothrombocytopenia complicated by bleeding. A small number of variants in GP1BA have been reported to cause a milder and dominant form of macrothrombocytopenia, but only 2 tentative reports exist of such a variant in GP1BB By analyzing data from a collection of more than 1000 genome-sequenced patients with a rare bleeding and/or platelet disorder, we have identified a significant association between rare monoallelic variants in GP1BB and macrothrombocytopenia. To strengthen our findings, we sought further cases in 2 additional collections in the United Kingdom and Japan. Across 18 families exhibiting phenotypes consistent with autosomal dominant inheritance of macrothrombocytopenia, we report on 27 affected cases carrying 1 of 9 rare variants in GP1BB., (© 2017 by The American Society of Hematology.)

Many common variants have been associated with hematological traits, but identification of causal genes and pathways has proven challenging. We performed a genome-wide association analysis in the UK Biobank and INTERVAL studies, testing 29.5 million genetic variants for association with 36 red cell, white cell, and platelet properties in 173,480 European-ancestry participants. This effort yielded hundreds of low frequency (<5%) and rare (<1%) variants with a strong impact on blood cell phenotypes. Our data highlight general properties of the allelic architecture of complex traits, including the proportion of the heritable component of each blood trait explained by the polygenic signal across different genome regulatory domains. Finally, through Mendelian randomization, we provide evidence of shared genetic pathways linking blood cell indices with complex pathologies, including autoimmune diseases, schizophrenia, and coronary heart disease and evidence suggesting previously reported population associations between blood cell indices and cardiovascular disease may be non-causal., (Copyright © 2016 Elsevier Inc. All rights reserved.)

Macrothrombocytopenia (MTP) is a heterogeneous group of disorders characterized by enlarged and reduced numbers of circulating platelets, sometimes resulting in abnormal bleeding. In most MTP, this phenotype arises because of altered regulation of platelet formation from megakaryocytes (MKs). We report the identification of DIAPH1, which encodes the Rho-effector diaphanous-related formin 1 (DIAPH1), as a candidate gene for MTP using exome sequencing, ontological phenotyping, and similarity regression. We describe 2 unrelated pedigrees with MTP and sensorineural hearing loss that segregate with a DIAPH1 R1213* variant predicting partial truncation of the DIAPH1 diaphanous autoregulatory domain. The R1213* variant was linked to reduced proplatelet formation from cultured MKs, cell clustering, and abnormal cortical filamentous actin. Similarly, in platelets, there was increased filamentous actin and stable microtubules, indicating constitutive activation of DIAPH1. Overexpression of DIAPH1 R1213* in cells reproduced the cytoskeletal alterations found in platelets. Our description of a novel disorder of platelet formation and hearing loss extends the repertoire of DIAPH1-related disease and provides new insight into the autoregulation of DIAPH1 activity., (© 2016 by The American Society of Hematology.)

Blood cells derive from hematopoietic stem cells through stepwise fating events. To characterize gene expression programs driving lineage choice, we sequenced RNA from eight primary human hematopoietic progenitor populations representing the major myeloid commitment stages and the main lymphoid stage. We identified extensive cell type-specific expression changes: 6711 genes and 10,724 transcripts, enriched in non-protein-coding elements at early stages of differentiation. In addition, we found 7881 novel splice junctions and 2301 differentially used alternative splicing events, enriched in genes involved in regulatory processes. We demonstrated experimentally cell-specific isoform usage, identifying nuclear factor I/B (NFIB) as a regulator of megakaryocyte maturation-the platelet precursor. Our data highlight the complexity of fating events in closely related progenitor populations, the understanding of which is essential for the advancement of transplantation and regenerative medicine., (Copyright © 2014, American Association for the Advancement of Science.)

Motivation: Most methods for estimating differential expression from RNA-seq are based on statistics that compare normalized read counts between treatment classes. Unfortunately, reads are in general too short to be mapped unambiguously to features of interest, such as genes, isoforms or haplotype-specific isoforms. There are methods for estimating expression levels that account for this source of ambiguity. However, the uncertainty is not generally accounted for in downstream analysis of gene expression experiments. Moreover, at the individual transcript level, it can sometimes be too large to allow useful comparisons between treatment groups., Results: In this article we make two proposals that improve the power, specificity and versatility of expression analysis using RNA-seq data. First, we present a Bayesian method for model selection that accounts for read mapping ambiguities using random effects. This polytomous model selection approach can be used to identify many interesting patterns of gene expression and is not confined to detecting differential expression between two groups. For illustration, we use our method to detect imprinting, different types of regulatory divergence in cis and in trans and differential isoform usage, but many other applications are possible. Second, we present a novel collapsing algorithm for grouping transcripts into inferential units that exploits the posterior correlation between transcript expression levels. The aggregate expression levels of these units can be estimated with useful levels of uncertainty. Our algorithm can improve the precision of expression estimates when uncertainty is large with only a small reduction in biological resolution., Availability and Implementation: We have implemented our software in the mmdiff and mmcollapse multithreaded C++ programs as part of the open-source MMSEQ package, available on https://github.com/eturro/mmseq.