A comparison of methods for differential expression analysis of RNA-seq data.

BACKGROUND: Finding genes that are differentially expressed between conditions is an integral part of understanding the molecular basis of phenotypic variation. In the past decades, DNA microarrays have been used extensively to quantify the abundance of mRNA corresponding to different genes, and more recently high-throughput sequencing of cDNA (RNA-seq) has emerged as a powerful competitor. As the cost of sequencing decreases, it is conceivable that the use of RNA-seq for differential expression analysis will increase rapidly. To exploit the possibilities and address the challenges posed by this relatively new type of data, a number of software packages have been developed especially for differential expression analysis of RNA-seq data. RESULTS: We conducted an extensive comparison of eleven methods for differential expression analysis of RNA-seq data. All methods are freely available within the R framework and take as input a matrix of counts, i.e. the number of reads mapping to each genomic feature of interest in each of a number of samples. We evaluate the methods based on both simulated data and real RNA-seq data. CONCLUSIONS: Very small sample sizes, which are still common in RNA-seq experiments, impose problems for all evaluated methods and any results obtained under such conditions should be interpreted with caution. For larger sample sizes, the methods combining a variance-stabilizing transformation with the 'limma' method for differential expression analysis perform well under many different conditions, as does the nonparametric SAMseq method.

Using graphical probability analysis (Bayes Nets) to evaluate a conditional DNA inclusion

This paper discusses the analysis of cases in which the inclusion or exclusion of a particular suspect, as a possible contributor to a DNA mixture, depends on the value of a variable (the number of contributors) that cannot be determined with certainty. It offers alternative ways to deal with such cases, including sensitivity analysis and object-oriented Bayesian networks, that separate uncertainty about the inclusion of the suspect from uncertainty about other variables. The paper presents a case study in which the value of DNA evidence varies radically depending on the number of contributors to a DNA mixture: if there are two contributors, the suspect is excluded; if there are three or more, the suspect is included; but the number of contributors cannot be determined with certainty. It shows how an object-oriented Bayesian network can accommodate and integrate varying perspectives on the unknown variable and how it can reduce the potential for bias by directing attention to relevant considerations and distinguishing different sources of uncertainty. It also discusses the challenge of presenting such evidence to lay audiences.

Differential sensitivity of GLUT1- and GLUT2-expressing beta cells to streptozotocin.

Streptozotocin injection in animals destroys pancreatic beta cells, leading to insulinopenic diabetes. Here, we evaluated the toxic effect of streptozotocin (STZ) in GLUT2(-/-) mice reexpressing either GLUT1 or GLUT2 in their beta cells under the rat insulin promoter (RIPG1 x G2(-/-) and RIPG2 x G2(-/-) mice, respectively). We demonstrated that injection of STZ into RIPG2 x G2(-/-) mice induced hyperglycemia (>20 mM) and an approximately 80% reduction in pancreatic insulin content. In vitro, the viability of RIPG2 x G2(-/-) islets was also strongly reduced. In contrast, STZ did not induce hyperglycemia in RIPG1 x G2(-/-) mice and did not reduce pancreatic insulin content. The viability of in vitro cultured RIPG1 x G2(-/-) islets was also unaffected by STZ. As islets from each type of transgenic mice were functionally indistinguishable, these data strongly support the notion that STZ toxicity toward beta cells depends on the expression of GLUT2.

Differential reactivity of TCR Vbeta10 alleles to a mouse mammary tumor virus superantigen.

Mouse mammary tumor virus (MMTV) expresses a superantigen (SAg) which plays a critical role in the viral life cycle. We have recently described the new infectious MMTV (SIM) encoding a Vbeta4-specific SAg in mice with a TCR-Vbeta(b) haplotype. We have now compared the SAg activity of this virus in BALB/c mice harboring the TCR-Vbeta(a), TCR-Vbeta(b) or TCR-Vbeta(c) haplotypes which differ by a central deletion in the TCR-Vbeta(a) and TCR-Vbeta(c) locus and by mutations in some of the remaining Vbeta elements. Injection of MMTV (SIM) led to a strong stimulation of Vbeta4+ CD4+ T cells in TCR-Vbeta(b) mice, but only to a weak stimulation of these cells in TCR-Vbeta(a) or TCR-Vbeta(c) mice. A large increase in the percentage of Vbeta10+ cells was observed among CD4+ T cells in mice with the Vbeta(a) or Vbeta(c), but not the Vbeta(b) TCR-Vbeta haplotype. Vbeta10+ cells dominated the response when Vbeta10(a/c) and Vbeta4 subsets were present together. This is the first report of a viral SAg interacting with murine Vbeta10+ cells. Six amino acid differences between Vbeta10(a/c) and Vbeta10(b) could account for the gain of reactivity of Vbeta10(a/c) to the MMTV(SIM) SAg. No mutations were found in the hypervariable region 4 (HV4) of the TCR. Mutations at positions 22 and 28 introduce into Vbeta10(a/c) the same amino acids which are found at these positions in the MMTV(SIM)-reactive Vbeta4. Tridimensional models indicated that these amino acids lie close to HV4 and are likely to be important for the interaction of the SAg with the TCR.

Differential effectiveness of microbially induced resistance against herbivorous insects in Arabidopsis.

Rhizobacteria-induced systemic resistance (ISR) and pathogen-induced systemic acquired resistance (SAR) have a broad, yet partly distinct, range of effectiveness against pathogenic microorganisms. Here, we investigated the effectiveness of ISR and SAR in Arabidopsis against the tissue-chewing insects Pieris rapae and Spodoptera exigua. Resistance against insects consists of direct defense, such as the production of toxins and feeding deterrents and indirect defense such as the production of plant volatiles that attract carnivorous enemies of the herbivores. Wind-tunnel experiments revealed that ISR and SAR did not affect herbivore-induced attraction of the parasitic wasp Cotesia rubecula (indirect defense). By contrast, ISR and SAR significantly reduced growth and development of the generalist herbivore S. exigua, although not that of the specialist P. rapae. This enhanced direct defense against S. exigua was associated with potentiated expression of the defense-related genes PDF1.2 and HEL. Expression profiling using a dedicated cDNA microarray revealed four additional, differentially primed genes in microbially induced S. exigua-challenged plants, three of which encode a lipid-transfer protein. Together, these results indicate that microbially induced plants are differentially primed for enhanced insect-responsive gene expression that is associated with increased direct defense against the generalist S. exigua but not against the specialist P. rapae.

Differential expression of peroxisome proliferator-activated receptor-alpha, -beta, and -gamma during rat embryonic development.

The expression patterns of the three different peroxisome proliferator-activated receptor (PPAR) isotypes have been determined during rat embryonic development by in situ hybridization. The expression of PPARalpha starts late in development, with increasing levels in organs such as liver, kidney, intestine, and pancreas, in which it will also be present later in adulthood to regulate its specific target genes. PPARalpha is also transiently expressed in the embryonic epidermis and central nervous system. PPARgamma presents a very restricted pattern of expression, being strongly expressed in brown adipose tissue, in which differentiation it has been shown to participate. Like PPARalpha, it is also expressed transiently in the central nervous system. Interestingly, PPARalpha, -beta and -gamma are coexpressed at high levels in brown adipose tissue. Finally, the high and ubiquitous expression of PPARbeta suggests some fundamental role(s) that this receptor might play throughout development.

Epigenetic modification of the FMR1 gene in fragile X syndrome is associated with differential response to the mGluR5 antagonist AFQ056.

Fragile X syndrome (FXS) is an X-linked condition associated with intellectual disability and behavioral problems. It is caused by expansion of a CGG repeat in the 5' untranslated region of the fragile X mental retardation 1 (FMR1) gene. This mutation is associated with hypermethylation at the FMR1 promoter and resultant transcriptional silencing. FMR1 silencing has many consequences, including up-regulation of metabotropic glutamate receptor 5 (mGluR5)-mediated signaling. mGluR5 receptor antagonists have shown promise in preclinical FXS models and in one small open-label study of FXS. We examined whether a receptor subtype-selective inhibitor of mGluR5, AFQ056, improves the behavioral symptoms of FXS in a randomized, double-blind, two-treatment, two-period, crossover study of 30 male FXS patients aged 18 to 35 years. We detected no significant effects of treatment on the primary outcome measure, the Aberrant Behavior Checklist-Community Edition (ABC-C) score, at day 19 or 20 of treatment. In an exploratory analysis, however, seven patients with full FMR1 promoter methylation and no detectable FMR1 messenger RNA improved, as measured with the ABC-C, significantly more after AFQ056 treatment than with placebo (P < 0.001). We detected no response in 18 patients with partial promoter methylation. Twenty-four patients experienced an adverse event, which was mostly mild to moderately severe fatigue or headache. If confirmed in larger and longer-term studies, these results suggest that blockade of the mGluR5 receptor in patients with full methylation at the FMR1 promoter may show improvement in the behavioral attributes of FXS.

Differential responses of newborn pulmonary arteries and veins to atrial and C-type natriuretic peptides.

Atrial natriuretic peptide (ANP) and C-type natriuretic peptide (CNP) are important dilators of the pulmonary circulation during the perinatal period. We compared the responses of pulmonary arteries (PA) and veins (PV) of newborn lambs to these peptides. ANP caused a greater relaxation of PA than of PV, and CNP caused a greater relaxation of PV than of PA. RIA showed that ANP induced a greater increase in cGMP content of PA than CNP. In PV, ANP and CNP caused a similar moderate increase in cGMP content. Receptor binding study showed more specific binding sites for ANP than for CNP in PA and more for CNP than for ANP in PV. Relative quantitative RT-PCR for natriuretic peptide receptor A (NPR-A) and B (NPR-B) mRNAs show that, in PA, NPR-A mRNA is more prevalent than NPR-B mRNA, whereas, in PV, NPR-B mRNA is more prevalent than NPR-A mRNA. In conclusion, in the pulmonary circulation, arteries are the major site of action for ANP, and veins are the major site for CNP. Furthermore, the differences in receptor abundance and the involvement of a cGMP-independent mechanism may contribute to the heterogeneous effects of the natriuretic peptides in PA and PV of newborn lambs.

Leishmania major: differential regulation of the surface metalloprotease in amastigote and promastigote stages.

During its life cycle, the protozoan parasite Leishmania major alternates from an intracellular amastigote form in the mammalian host to a flagellated promastigote form in the insect vector. The expression of the surface metalloprotease (PSP) during differentiation in vitro was investigated by Western and Northern blots, by immunoprecipitation of cells metabolically labeled with [35S]methionine or labeled at the surface with radioactive iodine, and by quantification of the proteolytic activity in substrate-containing polyacrylamide gels. We report that the surface metalloprotease is down-regulated at both the mRNA and the protein level in amastigotes, where it represents less than 1% of the equivalent proteolytic activity detected in promastigotes. A significant amount of mRNA is detected 4 hr after the onset of differentiation. The expression of the protease begins at that time and reaches steady state 8 hr later. The synthesis of PSP precedes the complete morphological differentiation to the promastigote stage and the appearance of the lipophosphoglycan, another major promastigote surface component. In contrast to PSP, a family of mercaptoethanol-activated proteases present in the amastigote exists only at a reduced level in the promastigote. The confinement of the surface metalloprotease to the insect stage of the parasite suggests that it has no physiological function in the parasitism maintenance of mammalian host macrophages.

Differential distribution of MAP1A isoforms in the adult mouse barrel cortex and comparison with the serotonin 5-HT2A receptor.

Microtubule-associated protein 1A (MAP1A) is essential during the late differentiation phase of neuronal development. Here, we demonstrated the presence of two MAP1A isoforms with a differential spatial distribution in the adult mouse barrel cortex. Antibody A stained MAP1A in pyramidal and stellate cells, including dendrites that crossed layer IV in the septa between barrels. The other antibody, BW6 recognized a MAP1A isoform that was mainly confined to the barrel hollow and identified smaller caliber dendrites. Previously, an interaction of MAP1A and the serotonin 5-hydroxytryptamine 2A (5-HT(2A)) receptor was shown in the rat cortex. Here, we identified, by double-immunofluorescent labeling, MAP1A isoform and serotonin 5-HT(2A) receptor distribution. MAP1A co-localized mainly with 5-HT(2A) receptor in larger apical dendrites situated in septa. This differential staining of MAP1A and a serotonin receptor in defined barrel compartments may be due to changes in the expression or processing of MAP1A during dendritic transport as a consequence of functional differences in processing of whisker-related sensory input.

Expression analyses of three members of the AtPHO1 family reveal differential interactions between signaling pathways involved in phosphate deficiency and the responses to auxin, cytokinin, and abscisic acid.

The PHO1 protein is involved in loading inorganic phosphate (Pi) to the root xylem. Ten genes homologous to AtPHO1 are present in the Arabidopsis thaliana (L.) Heyn genome. From this gene family, transcript levels of only AtPHO1, AtPHO1;H1 and AtPHO1;H10 were increased by Pi-deficiency. While the up-regulation of AtPHO1;H1 and AtPHO1;H10 by Pi deficiency followed the same rapid kinetics and was dependent on the PHR1 transcription factor, phosphite only strongly suppressed the expression of AtPHO1;H1 and had a minor effect on AtPHO1;H10. Addition of sucrose was found to increase transcript levels of both AtPHO1 and AtPHO1;H1 in Pi-sufficient or Pi-deficient plants, but to suppress AtPHO1:H10 under the same conditions. Treatments of plants with auxin or cytokinin had contrasting effect depending on the gene and on the Pi status of the plants. Thus, while both hormones down-regulated expression of AtPHO1 independently of the plant Pi status, auxin and cytokinin up-regulated AtPHO1;H1 and AtPHO1;H10 expression in Pi-sufficient plants and down-regulated expression in Pi-deficient plants. Treatments with abscisic acid inhibited AtPHO1 and AtPHO1;H1 expression in both Pi-sufficient and Pi-deficient plants, but increased AtPHO1;H10 expression under the same conditions. The inhibition of expression by abscisic acid of AtPHO1 and AtPHO1;H1, and of the Pi-starvation responsive genes AtPHT1;1 and AtIPS1, was dependant on the ABI1 type 2C protein phosphatase. These results reveal that various levels of cross talk between the signal transduction pathways to Pi, sucrose and phytohormones are involved in the regulation of expression of the three AtPHO1 homologues.

Differential expression and hypoxic regulation of adenosine receptors and TRPC channels in the developing heart.

Objectives: To characterize the modifications of gene expression of adenosine receptors (AR), TRPC channels, HIF-1α and iNOS during the early cardiogenesis in response to chronic hypoxia exposure. Methods: 4-day-old chick embryos were subjected in ovo to 6H, 12H and 24H of hypoxia (10% O2). The mRNA expression was quantified by RT-qPCR. Results: The targeted genes were found to be expressed at mRNA level with a differential expression pattern within the heart. Hypoxia has no significant effect on mRNA expression of ARs, TRPCs channels and iNOS within the heart. By contrast, HIF-1α mRNA expression shows a tendency to be down-regulated by hypoxia. Conclusion: These results suggest that an intrauterine oxygen lack does not significantly affect expression of genes involved in adenosine signaling and in calcium handling by store operated channels (TRPC).