A new method for class prediction based on signed-rank algorithms applied to Affymetrix<Superscript>®</Superscript> microarray experiments

Background

The huge amount of data generated by DNA chips is a powerful basis to classify various pathologies. However, constant evolution of microarray technology makes it difficult to mix data from different chip types for class prediction of limited sample populations. Affymetrix^® technology provides both a quantitative fluorescence signal and a decision (detection call: absent or present) based on signed-rank algorithms applied to several hybridization repeats of each gene, with a per-chip normalization. We developed a new prediction method for class belonging based on the detection call only from recent Affymetrix chip type. Biological data were obtained by hybridization on U133A, U133B and U133Plus 2.0 microarrays of purified normal B cells and cells from three independent groups of multiple myeloma (MM) patients.

Results

After a call-based data reduction step to filter out non class-discriminative probe sets, the gene list obtained was reduced to a predictor with correction for multiple testing by iterative deletion of probe sets that sequentially improve inter-class comparisons and their significance. The error rate of the method was determined using leave-one-out and 5-fold cross-validation. It was successfully applied to (i) determine a sex predictor with the normal donor group classifying gender with no error in all patient groups except for male MM samples with a Y chromosome deletion, (ii) predict the immunoglobulin light and heavy chains expressed by the malignant myeloma clones of the validation group and (iii) predict sex, light and heavy chain nature for every new patient. Finally, this method was shown powerful when compared to the popular classification method Prediction Analysis of Microarray (PAM).

Conclusion

This normalization-free method is routinely used for quality control and correction of collection errors in patient reports to clinicians. It can be easily extended to multiple class prediction suitable with clinical groups, and looks particularly promising through international cooperative projects like the "Microarray Quality Control project of US FDA" MAQC as a predictive classifier for diagnostic, prognostic and response to treatment. Finally, it can be used as a powerful tool to mine published data generated on Affymetrix systems and more generally classify samples with binary feature values.

     

Towards biocompatible vaccine delivery systems: interactions of colloidal PECs based on polysaccharides with HIV-1 p24 antigen

This work reports on the interactions of a model protein (p24, the capside protein of HIV-1 virus) with colloids obtained from polyelectrolyte complexes (PECs) involving two polysaccharides: chitosan and dextran sulfate (DS). The PECs were elaborated by a one-shot addition of default amounts of one counterpart to the polymer in excess. Depending on the nature of the excess polyelectrolyte, the submicrometric colloid was either positively or negatively charged. HIV-1 capsid p24 protein was chosen as antigen, the ultrapure form, lipopolysaccharide-free (endotoxin-, vaccine grade) was used in most experiments, as the level of purity of the protein had a great impact on the immobilization process. p24 sorption kinetics, isotherms, and loading capacities were investigated for positively and negatively charged particles of chitosans and dextran sulfates differing in degrees of polymerization (DP) or acetylation (DA). Compared with the positive particles, negatively charged colloids had higher binding capacities, faster kinetics, and a better stability of the adsorbed p24. Capacities up to 600 mg x g(-1) (protein-colloid) were obtained, suggesting that the protein interacted within the shell of the particles. Small-angle X-rays scattering experiments confirmed this hypothesis. Finally, the immunogenicity of the p24-covered particles was assessed for vaccine purposes in mice. The antibody titers obtained with immobilized p24 was dose dependent and in the same range as for Freund's adjuvant, a gold standard for humoral responses.     

CD200: a putative therapeutic target in cancer

CD200 was recently described as a new prognosis factor in multiple myeloma and acute myeloid leukemia. CD200 is a membrane glycoprotein that imparts an immunoregulatory signal through CD200R, leading to the suppression of T-cell-mediated immune responses. We investigated the expression of CD200 in cancer using publicly available gene expression data. CD200 gene expression in normal or malignant human tissues or cell lines was obtained from the Oncomine Cancer Microarray database, Amazonia database and the ITTACA database. We found significant overexpression of CD200 in renal carcinoma, head and neck carcinoma, testicular cancer, malignant mesothelioma, colon carcinoma, MGUS/smoldering myeloma, and in chronic lymphocytic leukemia compared to their normal cells or their tissue counterparts. Moreover, we show that CD200 expression is associated with tumor progression in various cancers. Taken together, these data suggest that CD200 is a potential therapeutic target and prognostic factor for a large array of malignancies.     

Molecular hydrogen from water radiolysis as an energy source for bacterial growth in a basin containing irradiating waste

Although being deionized, filtered and therefore normally deeply oligotrophic, the water from a basin containing irradiating waste presented relatively high bacterial concentrations (ca 10(5) cfu ml(-1)) and biofilm development at its surface and on the walls. This water was characterized by a high concentration of molecular H2 due to water radiolysis, while its electrochemical potential was around +400 mV due the presence of dissolved O2 and active oxygen compounds. This combination of H2 availability and of an oxidant environment is completely original and not described in nature. From surface and wall biofilms, we enumerated the autotrophic populations ( approximately 10(5) bacteria ml(-1)) able to grow in presence of H2 as energy source and CO2 as carbon source, and we isolated the most abundant ones among cultivable bacteria. They efficiently grew on a mineral medium, in the presence of H2, O2 and CO2, the presence of the three gases being indispensable. Two strains were selected and identified using their rrs gene sequence as Ralstonia sp. GGLH002 and Burkholderia sp. GGLH005. In pure culture and using isotope exchange between hydrogen and deuterium, we demonstrated that these strains are able to oxidize hydrogen as energy source, using oxygen as an electron acceptor, and to use carbon dioxide as carbon source. These chemoautotroph hydrogen-oxidizing bacteria probably represent the pioneer bacterial populations in this basin and could be primary producers in the bacterial community.     

RAB-5 controls the cortical organization and dynamics of PAR proteins to maintain C. elegans early embryonic polarity

In all organisms, cell polarity is fundamental for most aspects of cell physiology. In many species and cell types, it is controlled by the evolutionarily conserved PAR-3, PAR-6 and aPKC proteins, which are asymmetrically localized at the cell cortex where they define specific domains. While PAR proteins define the antero-posterior axis of the early C. elegans embryo, the mechanism controlling their asymmetric localization is not fully understood. Here we studied the role of endocytic regulators in embryonic polarization and asymmetric division. We found that depleting the early endosome regulator RAB-5 results in polarity-related phenotypes in the early embryo. Using Total Internal Reflection Fluorescence (TIRF) microscopy, we observed that PAR-6 is localized at the cell cortex in highly dynamic puncta and depleting RAB-5 decreased PAR-6 cortical dynamics during the polarity maintenance phase. Depletion of RAB-5 also increased PAR-6 association with clathrin heavy chain (CHC-1) and this increase depended on the presence of the GTPase dynamin, an upstream regulator of endocytosis. Interestingly, further analysis indicated that loss of RAB-5 leads to a disorganization of the actin cytoskeleton and that this occurs independently of dynamin activity. Our results indicate that RAB-5 promotes C. elegans embryonic polarity in both dynamin-dependent and -independent manners, by controlling PAR-6 localization and cortical dynamics through the regulation of its association with the cell cortex and the organization of the actin cytoskeleton.     

Differential expression of sarcoplasmic proteins in four heterogeneous ovine skeletal muscles

Fiber-type distribution is known to vary widely within and between muscles according to differences in muscle functions. 2-DE and MALDI-MS were used to investigate the molecular basis of muscle fiber type-related variability. We compared four lamb skeletal muscles with heterogeneous fiber-type composition that are relatively rich in fast-twitch fiber types, i.e., the semimembranosus, vastus medialis, longissimus dorsi, and tensor fasciae latae (TL). Our results clearly showed that none of the glycolytic metabolism enzymes detected, including TL which was most strongly glycolytic, made intermuscular differentiation possible. Muscle differentiation was based on the differential expression of proteins involved in oxidative metabolism, including not only citric acid cycle enzymes but also other classes of proteins with functions related to oxidative metabolism, oxidative stress, and probably to higher protein turnover. Detected proteins were involved in transport (carbonate dehydratase, myoglobin, fatty acid-binding protein), repair of misfolding damage (heat shock protein (HSP) 60 kDa, HSP-27 kDa, alpha-crystallin beta subunit, DJ1, stress-induced phosphoprotein), detoxification or degradation of impaired proteins (GST-Pi, aldehyde dehydrogenase, peroxiredoxin, ubiquitin), and protein synthesis (tRNA-synthetase). The fractionating method led to the detection of proteins involved in different functions related to oxidative metabolism that have not previously been shown concomitancy.     

Mixing Languages during Learning? Testing the One Subject—One Language Rule

In bilingual communities, mixing languages is avoided in formal schooling: even if two languages are used on a daily basis for teaching, only one language is used to teach each given academic subject. This tenet known as the one subject-one language rule avoids mixing languages in formal schooling because it may hinder learning. The aim of this study was to test the scientific ground of this assumption by investigating the consequences of acquiring new concepts using a method in which two languages are mixed as compared to a purely monolingual method. Native balanced bilingual speakers of Basque and Spanish—adults (Experiment 1) and children (Experiment 2)—learnt new concepts by associating two different features to novel objects. Half of the participants completed the learning process in a multilingual context (one feature was described in Basque and the other one in Spanish); while the other half completed the learning phase in a purely monolingual context (both features were described in Spanish). Different measures of learning were taken, as well as direct and indirect indicators of concept consolidation. We found no evidence in favor of the non-mixing method when comparing the results of two groups in either experiment, and thus failed to give scientific support for the educational premise of the one subject—one language rule.     

Genetics of frost hardiness in <Emphasis Type="Italic">Juglans regia</Emphasis> L. and relationship with growth and phenology

The growing interest in broadleaf timber plantations in the Mediterranean area has promoted several studies focusing on the identification and characterization of variability sources in main timber-producing species. J. regia is one of these species, well-adapted to this area, but with freezing, damages registrations. Breeding focused on productive traits should include knowledge of adaptation, required to obtain a good selection capable of producing a suitable turnover in timber plantations. In this study, the features evaluated were autumn and winter frost hardiness and some vegetative traits on 22 half-sib J. regia progenies. Budsticks were exposed to sub-zero temperatures in a controlled chamber and using measurements of relative electrolyte content, the LT50 values (°C) were calculated by each individual. The study was carried out on seven-year-old progenies. The familiar heritability of autumn frost hardiness was 0.68, and on winter, it was 0.77. The autumn frost behaviour correlated genetically with the length of the growing season (0.574 ± 0.351), and both autumn and winter frost hardiness correlated inversely with secondary annual growth measured at breast height (?0.654 ± 0.259 and ?0.740 ± 0.227, respectively). These results pointed that growth could therefore be improved without increasing the frost vulnerability. This should be important for growers, particularly under climate change conditions.     

Peptidomic analysis of skin secretions from Rana heckscheri and Rana okaloosae provides insight into phylogenetic relationships among frogs of the Aquarana species group

The members of the Aquarana (or Rana catesbeiana species group) form a monophyletic group comprising seven species: R. catesbeiana, Rana clamitans, Rana grylio, Rana virgatipes, Rana septentrionalis, Rana heckscheri and Rana okaloosae. Previous work has led to structural characterization of the antimicrobial peptides present in electrically-stimulated skin secretions from the first five species listed and this study presents the primary structures of orthologs from the river frog R. heckscheri and the Florida bog frog R. okaloosae. Peptidomic analysis of R. heckscheri and R. okaloosae skin secretions led to the identification of peptides with antimicrobial activity belonging to the ranalexin, ranatuerin-2, and temporin families. In addition, a peptide (GFLDIIKDTGKDFAVKILNNLKCKLAGGCPR) was isolated from R. okaloosae whose primary structure identified it as a member of the palustrin-2 family. Consistent with previous data based upon morphological analysis and comparisons of the nucleotide sequences of mitochondrial and ribosomal genes, cladistic analysis based upon a comparison of the amino acid sequences of antimicrobial peptides indicates a sister-group relationship between R. heckscheri and R. grylio and a close, but less well defined, phylogenetic relationship between R. okaloosae and R. clamitans.     

Comparison of different neuron models to conductance-based post-stimulus time histograms obtained in cortical pyramidal cells using dynamic-clamp in vitro

A wide diversity of models have been proposed to account for the spiking response of central neurons, from the integrate-and-fire (IF) model and its quadratic and exponential variants, to multiple-variable models such as the Izhikevich (IZ) model and the well-known Hodgkin–Huxley (HH) type models. Such models can capture different aspects of the spiking response of neurons, but there is few objective comparison of their performance. In this article, we provide such a comparison in the context of well-defined stimulation protocols, including, for each cell, DC stimulation, and a series of excitatory conductance injections, arising in the presence of synaptic background activity. We use the dynamic-clamp technique to characterize the response of regular-spiking neurons from guinea-pig visual cortex by computing families of post-stimulus time histograms (PSTH), for different stimulus intensities, and for two different background activities (low- and high-conductance states). The data obtained are then used to fit different classes of models such as the IF, IZ, or HH types, which are constrained by the whole data set. This analysis shows that HH models are generally more accurate to fit the series of experimental PSTH, but their performance is almost equaled by much simpler models, such as the exponential or pulse-based IF models. Similar conclusions were also reached by performing partial fitting of the data, and examining the ability of different models to predict responses that were not used for the fitting. Although such results must be qualified by using more sophisticated stimulation protocols, they suggest that nonlinear IF models can capture surprisingly well the response of cortical regular-spiking neurons and appear as useful candidates for network simulations with conductance-based synaptic interactions.