Nonparametric Bayesian sparse factor models with application to gene expression modeling

A nonparametric Bayesian extension of Factor Analysis (FA) is proposed where observed data $\mathbf{Y}$ is modeled as a linear superposition, $\mathbf{G}$, of a potentially infinite number of hidden factors, $\mathbf{X}$. The Indian Buffet Process (IBP) is used as a prior on $\mathbf{G}$ to incorporate sparsity and to allow the number of latent features to be inferred. The model's utility for modeling gene expression data is investigated using randomly generated data sets based on a known sparse connectivity matrix for E. Coli, and on three biological data sets of increasing complexity.

MicroRNA regulation and the variability of human cortical gene expression

Understanding the driving forces of gene expression variation within human populations will provide important insights into the molecular basis of human phenotypic variation. In the genome, the gene expression variability differs among genes, and at prese

Embryonic stages and eye-specific gene expression of the local cyprinoid fish Anabarilius grahami in Fuxian Lake, China

Anabarilius grahami is a cyprinoid fish endemic to Fuxian Lake, Yunnan, China. In this study, a comprehensive staging series of A. grahami was produced. The embryonic development of A. grahami was divided into six main periods: zygote period, cleavage per

Some biocybernetic aspects in gene expression and regulatory control

In eucaryotes, gene expression and control is a complex nonlinear process, where there are many control mechanisms and ways, both physic, chemical and informational control. By the exploration from the angle of biocybernetics, the authors suggest that gene expression is a co-control process. In this process, physic, chemical and informational feedback controls are associated and influential each other, and are cross and co-functional. The physic, chemical and informational control ways composed an order non-linear feedback control system in eucaryotes.

Effects of exogenous double-stranded RNA on the basonuclin gene expression in mouse oocytes

In plants and less-advanced animal species, such as C.elegans, introduction of exogenous double-stranded RNA (dsRNA) into cells would trigger degradation of the mRNA with homologous sequence and interfere with the endogenous gene expression. It might represent an ancient anti-virus response which could prevent the mutation in the genome that was caused by virus infection or mobile DNA elements insertion. This phenomenon was named RNA interference, or RNAi. In this study, RNAi was used to investigate the function of basonuclin gene during oogenesis. Microinjection of dsRNA directed towards basonuclin into mouse germinal-vesicle-intact (GV) oocytes brought down the abundance of the cognate mRNA effectively in a time- and concentration-dependent manner. This reduction effect was sequence-specific and showed no negative effect on other non-homologous gene expression in oocytes, which indicated that dsRNA can recognize and cause the degradation of the transcriptional products of endogenous basonuclin gene in a sequence-specific manner. Immunofluorescence results showed that RNAi could reduce the concentration of basonuclin protein to some extent, but the effect was less efficient than the dsRNA targeting towards tPA and cMos which was also expressed in oocytes. This result might be due to the long half life of basonuclin protein in oocytes and the short reaction time which was posed by the limited life span of GV oocytes cultured in vitro. In summary, dsRNA could inhibit the expression of the cognate gene in oocytes at both mRNA and protein levels. The effect was similar to Knock-out technique which was based on homologous recombination. Furthermore, hairpin-style dsRNA targeting basonuclin gene could be produced by transcription from a recombinant plasmid and worked efficiently to deplete the cognate mRNA in oocytes. This finding offered a new way to study the function of basonuclin in the early stage of oogenesis by infection of primordial oocytes with the plasmid expressing hairpin-style basonuclin dsRNA.

Regulation of zygotic gene activation in the preimplantation mouse embryo: Global activation and repression of gene expression

Superimposed on the activation of the embryonic genome in the preimplantation mouse embryo is the formation of a transcriptionally repressive state during the two-cell stage. This repression appears mediated at the level of chromatin structure, because it is reversed by inducing histone hyperacetylation or inhibiting the second round of DNA replication. We report that of more than 200 amplicons analyzed by mRNA differential display, about 45% of them are repressed between the two-cell and four-cell stages. This repression is scored as either a decrease in amplicon expression that occurs between the two-cell and four-cell stages or on the ability of either trichostatin A tan inhibitor of histone deacetylases) or aphidicolin tan inhibitor of replicative DNA polymerases) to increase the level of amplicon expression. Results of this study also indicate that about 16% of the amplicons analyzed likely are novel genes whose sequence doesn't correspond to sequences in the current databases, whereas about 20% of the sequences expressed during this transition likely are repetitive sequences. Lastly, inducing histone hyperacetylation in the two-cell embryos inhibits cleavage to the four-cell stage. These results suggest that genome activation is global and relatively promiscuous and that a function of the transcriptionally repressive state is to dictate the appropriate profile of gene expression that is compatible with further development.

Geometric optimization methods for independent component analysis applied on gene expression data

DNA microarrays provide a huge amount of data and require therefore dimensionality reduction methods to extract meaningful biological information. Independent Component Analysis (ICA) was proposed by several authors as an interesting means. Unfortunately, experimental data are usually of poor quality- because of noise, outliers and lack of samples. Robustness to these hurdles will thus be a key feature for an ICA algorithm. This paper identifies a robust contrast function and proposes a new ICA algorithm. © 2007 IEEE.

Effect of beta-glucan on activity of antioxidant enzymes and Mx gene expression in virus infected grass carp

The effects of beta-glucan, an immunostimulatory agent, on the superoxide dismutase (SOD) and catalase (CAT) activities of erythrocytes and Mx gene expression were studied from grass carp that were challenged with grass carp hemorrhage virus (GCHV). The SOD and CAT activities in erythrocytes and Mx gene expression in spleen from the fish were detected by spectrophotometry and RT-PCR, respectively. Negative control fish were injected with PBS; positive control groups were injected with either P-glucan or GCHV only; and the experimental groups were pre-injected with beta-glucan 15 days prior to injection with GCHV. The results show that the SOD and CAT activities were higher in fish injected with beta-glucan for 15 days than the negative control group injected with PBS. The SOD and CATactivities significantly decreased when the fish were challenged with GCHV, but it was higher in the group pre-treated with beta-glucan than in infected fish not pre-treated, 15 days after GCHV infection. Mx gene expression levels increased during the early stages (at 12 h and 36 h) of GCHV infection, and it remained at higher levels from the 6th till the 10th day in the beta-glucan pre-treated group, but it was failing from the 6th day in the beta-glucan untreated group. The GCHV-infected group pre-treated with P-glucan had a higher survival rate (60%) than the group not pre-treated with P-glucan (20%), suggesting that beta-glucan possesses or enhances anti-viral responses. (C) 2009 Elsevier Ltd. All rights reserved.