Evaluating the Impact of Prior Assumptions in Bayesian Biostatistics

A common concern in Bayesian data analysis is that an inappropriately informative prior may unduly influence posterior inferences. In the context of Bayesian clinical trial design, well chosen priors are important to ensure that posterior-based decision rules have good frequentist properties. However, it is difficult to quantify prior information in all but the most stylized models. This issue may be addressed by quantifying the prior information in terms of a number of hypothetical patients, i.e., a prior effective sample size (ESS). Prior ESS provides a useful tool for understanding the impact of prior assumptions. For example, the prior ESS may be used to guide calibration of prior variances and other hyperprior parameters. In this paper, we discuss such prior sensitivity analyses by using a recently proposed method to compute a prior ESS. We apply this in several typical settings of Bayesian biomedical data analysis and clinical trial design. The data analyses include cross-tabulated counts, multiple correlated diagnostic tests, and ordinal outcomes using a proportional-odds model. The study designs include a phase I trial with late-onset toxicities, a phase II trial that monitors event times, and a phase I/II trial with dose-finding based on efficacy and toxicity.     

The REDUCED LEAFLET Genes Encode Key Components of the trans-Acting Small Interfering RNA Pathway and Regulate Compound Leaf and Flower Development in Lotus japonicus

The endogenous trans-acting small interfering RNA (ta-siRNA) pathway plays a conserved role in adaxial-abaxial patterning of lateral organs in simple-leafed plant species. However, its function in compound-leafed species is largely unknown. Using the compound-leafed species Lotus japonicus, we identified and characterized two independent mutants, reduced leaflet1 (rel1) and rel3, whose most conspicuous defects in compound leaves are abaxialized leaflets and reduction in leaflet number. Concurrent mutations in REL genes also compromise flower development and result in radial symmetric floral organs. Positional cloning revealed that REL1 and REL3 encode the homologs of Arabidopsis (Arabidopsis thaliana) SUPPRESSOR OF GENE SILENCING3 and ARGONAUTE7/ZIPPY, respectively, which are key components of the ta-siRNA pathway. These observations, together with the expression and functional data, demonstrated that the ta-siRNA pathway plays conserved yet distinct roles in the control of compound leaf and flower development in L. japonicus. Moreover, the phenotypic alterations of lateral organs in ta-siRNA-deficient mutants and the regulation of downstream targets by the ta-siRNA pathway in L. japonicus were similar to those in the monocots but different from Arabidopsis, indicating many parallels between L. japonicus and the monocots in the control of lateral organ development by the ta-siRNA pathway.Plant endogenous small RNAs can be categorized into microRNAs (miRNAs) and small interfering RNAs (siRNAs) according to their mechanism of biogenesis (Vaucheret, 2006). trans-Acting siRNAs (ta-siRNAs) are one type of siRNA, and their biogenesis requires several key components, such as SUPPRESSOR OF GENE SILENCING3 (SGS3), RNA-DEPENDENT RNA POLYMERASE6 (RDR6), DICER-LIKE4 (DCL4), ARGONAUTE7 (AGO7)/ZIPPY (ZIP), and dsRNA-BINDING4 (Peragine et al., 2004; Vazquez et al., 2004; Gasciolli et al., 2005; Xie et al., 2005; Yoshikawa et al., 2005; Adenot et al., 2006; Nakazawa et al., 2007). Recent studies revealed that the ta-siRNA pathway is integrated into different processes of plant development, such as vegetative phase transition in Arabidopsis (Arabidopsis thaliana; Hunter et al., 2003; Peragine et al., 2004; Xie et al., 2005; Nakazawa et al., 2007) and shoot apical meristem (SAM) initiation in rice (Oryza sativa; Satoh et al., 1999; Itoh et al., 2000; Nagasaki et al., 2007). Parallel studies of this pathway in simple-leafed species also showed that the ta-siRNA pathway plays critical roles in patterning of leaves and floral organs.In flowering plants, leaves and flowers are produced on the periphery of the apical meristem. These lateral organs are structurally asymmetric with regard to the apical meristem. The adaxial side is adjacent to the meristem, while the abaxial side is away from the meristem. The ta-siRNA pathway was found to play a conserved role in specifying the adaxial identity of lateral organs in both monocots and dicots, but defects in the ta-siRNA pathway caused more severe phenotypes in monocots than in dicot Arabidopsis. In Arabidopsis, no clear leaf polarity defects were detected in the ta-siRNA-defective mutants. However, blocking the ta-siRNA pathway in asymmetric1 (as1) or as2 background, which are regulators of leaf adaxial identity (Lin et al., 2003; Xu et al., 2003), results in enhanced adaxial-abaxial leaf defects (Li et al., 2005; Xu et al., 2006; Garcia et al., 2006). In addition, the as2rdr6 double mutants also display aberrant flowers with sepals failing to enwrap the inner whorl organs and some sepals and petals becoming needle-like structures (Li et al., 2005). In maize (Zea mays), mutations in LEAFBLADELESS1 (LBL1), which encodes the Arabidopsis SGS3 ortholog, give rise to abnormal leaves with partial or complete loss of adaxial cell identity (Timmermans et al., 1998; Nogueira et al., 2007). In severe lbl1 mutants, leaf-like lateral organs of inflorescences and flowers develop as symmetric, thread-like organs, and the immature ear is exposed and arrested in development (Timmermans et al., 1998). In rice, the osdcl4-1 mutants display an abaxialized epidermis in coleoptiles and in the first leaf, and knockdown of OsDCL4 can lead to the awn-like lemma with a radial abaxialized identity and the stamens and carpel not enwrapped by the lemma and pelea (Liu et al., 2007). Transgenic rice plants with ectopic expression of SHOOTLESS4 (SHL4), the homolog of Arabidopsis AGO7, exhibit partially adaxialized leaves (Nagasaki et al., 2007; Shi et al., 2007).In addition to the ta-siRNA pathway, other components have also been shown to be involved in the adaxial-abaxial patterning of lateral organs. The Antirrhinum majus PHANTASTICA (PHAN) gene (Waites et al., 1998; Byrne et al., 2000; Xu et al., 2003; Qi et al., 2004), which is the ortholog of Arabidopsis AS1, and CLASS III HOMEODOMAIN-LEUCINE ZIPPER (HD-ZIP III) gene family members (McConnell et al., 2001; Emery et al., 2003) contribute to adaxial pattern formation of lateral organs, whereas members of YABBY (YAB; Sawa et al., 1999; Siegfried et al., 1999) and KANADI (Eshed et al., 2001; Kerstetter et al., 2001) gene families, AUXIN RESPONSE FACTOR3 (ARF3) and ARF4 (Pekker et al., 2005), and the miRNAs miR165/166 (Emery et al., 2003; Eshed et al., 2004; Mallory et al., 2004) are required for specifying abaxial identity. How the activities of these adaxial and abaxial determinants are coordinated has been extensively studied. It was found that ARF3 and ARF4 are regulated by the TAS3 ta-siRNA, and this regulation is conserved in both monocots and dicots (Allen et al., 2005; Williams et al., 2005). Recent studies in Arabidopsis suggest that ta-siRNAs act in a non-cell-autonomous manner to spatially restrict ARF activity (Chitwood et al., 2009; Schwab et al., 2009).In contrast to simple leaves with their single lamina, compound leaves are composed of one petiole and several leaflets. It is found that genes required for the adaxial-abaxial patterning of lateral organs in simple-leafed species also play critical roles in compound-leafed species, but these genes play multiple roles in compound leaf development. In tomato (Solanum lycopersicum), down-regulation of PHAN ortholog disturbs the leaf polarity as well as leaflet formation (Kim et al., 2003). Extensive studies of the PHAN expression in diverse compound-leafed species suggest that the function of PHAN in maintaining leaf adaxial identity is associated with leaflet formation in compound leaves and reduced adaxial identity of leaf primordia by down-regulation of PHAN could change pinnate compound leaves into palmate leaves (Kim et al., 2003). In pea (Pisum sativum), the role of PHAN in compound leaf development has also been elucidated by characterization of the phan mutant crispa (cri; Tattersall et al., 2005). However, unlike antisense PHAN transgenic tomato leaves, the cri mutant has the individual leaflet abaxialized, rather than the whole leaf. The number of lateral organs on the cri mutant compound leaves, including leaflets, is not altered, and the leaves remain pinnate. Apart from leaf development, the cri mutation also affects flower development. Although the floral organ identity and organ number are not altered, the laminar floral organ display abaxialized identity (Tattersall et al., 2005).The ta-siRNA pathway plays a critical role in simple-leafed species, but its role in compound-leafed species is not understood. Here, we address this question by analyzing loss-of-function reduced leaflet (rel1) and rel3 mutants in the compound-leafed species Lotus japonicus. Phenotypic characterization shows compound leaves of rel mutants exhibit a conspicuous disturbance in leaflet polarity as well as reduction in leaflet number. Besides the abnormal compound leaves, flower development is also severely affected in rel mutants, showing radial symmetric petals. REL1 and REL3 were identified by map-based cloning and were shown to be homologs of Arabidopsis SGS3 and AGO7, respectively. REL1 and REL3 act in the same genetic pathway and are both required for the biogenesis of TAS3 ta-siRNA. Further investigation reveals that the homolog of the Arabidopsis ARF3 is duplicated in the L. japonicus genome and that the duplicate ARF3 homologs and the ARF4 homolog are all negatively regulated by the ta-siRNA pathway. Furthermore, we found that the expression of LjYAB1, a homolog of Arabidopsis YAB1, was decreased in rel mutants, which may be associated with the reduced lamina.Taken together, our data reveal that the ta-siRNA pathway is integrated into the regulatory networks in the control of lateral organ development in L. japonicus and further emphasize the importance of the ta-siRNA pathway in compound leaf development. Moreover, our results also indicate many parallels between L. japonicus and monocots for the ta-siRNA pathway in the regulation of lateral organs.     

Differential expression of Tenomodulin and Chondromodulin-1 at the insertion site of the tendon reflects a phenotypic transition of the resident cells

The insertion site of the tendon to the skeletal element is hypovascular and is one of the most common sites of dysfunction in the musculoskeletal system. However, the resident cells have been poorly defined due to a lack of a specific marker for tenocytes. We previously reported that Tenomodulin (Tnmd) and Chondromodulin-1 (Chm1) are homologous angiogenesis inhibitors and predominantly expressed in the avascular region of tendons and cartilage, respectively. In this study, we analyzed the expression of Tnmd, Chm1, alpha 1 chain of the type I collagen (Col1a1) and alpha 1 chain of the type II collagen (Col2a1) at the insertion site of the Achilles, patellar, or rotator cuff tendons of 1-week-old rabbits by in situ hybridization analysis. Tnmd was co-expressed with Col1a1 in tenocytes of these tendons, while Chm1 and Col2a1 were detected in chondrocytes of the hyaline cartilage. Interestingly, the cell population between Tnmd/Col1a1 positive tenocytes and Chm1/Col2a1 positive chondrocytes expressed Col1a1 but none of the other markers (Tnmd, Chm1, and Col2a1). Red blood cells were exclusively present at the interface between the tendon substance and cartilage in the insertion site of the Achilles tendon. Lack of Tnmd and Chm1 in this newly characterized cell population may allow the transitional zone between the poorly vascularized tendon and cartilage to establish the unique vascular pattern for blood supply.     

Six RNA Viruses and Forty-One Hosts: Viral Small RNAs and Modulation of Small RNA Repertoires in Vertebrate and Invertebrate Systems

We have used multiplexed high-throughput sequencing to characterize changes in small RNA populations that occur during viral infection in animal cells. Small RNA-based mechanisms such as RNA interference (RNAi) have been shown in plant and invertebrate systems to play a key role in host responses to viral infection. Although homologs of the key RNAi effector pathways are present in mammalian cells, and can launch an RNAi-mediated degradation of experimentally targeted mRNAs, any role for such responses in mammalian host-virus interactions remains to be characterized. Six different viruses were examined in 41 experimentally susceptible and resistant host systems. We identified virus-derived small RNAs (vsRNAs) from all six viruses, with total abundance varying from “vanishingly rare” (less than 0.1% of cellular small RNA) to highly abundant (comparable to abundant micro-RNAs “miRNAs”). In addition to the appearance of vsRNAs during infection, we saw a number of specific changes in host miRNA profiles. For several infection models investigated in more detail, the RNAi and Interferon pathways modulated the abundance of vsRNAs. We also found evidence for populations of vsRNAs that exist as duplexed siRNAs with zero to three nucleotide 3′ overhangs. Using populations of cells carrying a Hepatitis C replicon, we observed strand-selective loading of siRNAs onto Argonaute complexes. These experiments define vsRNAs as one possible component of the interplay between animal viruses and their hosts.