Integrative omics-analysis of lipid metabolism regulation by peroxisome proliferator-activated receptor a and b agonists in male Atlantic cod.

Lipid metabolism is essential in maintaining energy homeostasis in multicellular organisms. In vertebrates, the peroxisome proliferator-activated receptors (PPARs, NR1C) regulate the expression of many genes involved in these processes. Atlantic cod (Gadus morhua) is an important fish species in the North Atlantic ecosystem and in human nutrition, with a highly fatty liver. Here we study the involvement of Atlantic cod Ppar a and b subtypes in systemic regulation of lipid metabolism using two model agonists after in vivo exposure. WY-14,643, a specific PPARA ligand in mammals, activated cod Ppara1 and Ppara2 in vitro. In vivo, WY-14,643 caused a shift in lipid transport both at transcriptional and translational level in cod. However, WY-14,643 induced fewer genes in the fatty acid beta-oxidation pathway compared to that observed in rodents. Although GW501516 serves as a specific PPARB/D ligand in mammals, this compound activated cod Ppara1 and Ppara2 as well as Pparb in vitro. In vivo, it further induced transcription of Ppar target genes and caused changes in lipid composition of liver and plasma. The integrative approach provide a foundation for understanding how Ppars are engaged in regulating lipid metabolism in Atlantic cod physiology. We have shown that WY-14,643 and GW501516 activate Atlantic cod Ppara and Pparb, affect genes in lipid metabolism pathways, and induce changes in the lipid composition in plasma and liver microsomal membranes. Particularly, the combined transcriptomic, proteomics and lipidomics analyses revealed that effects of WY-14,643 on lipid metabolism are similar to what is known in mammalian studies, suggesting conservation of Ppara functions in mediating lipid metabolic processes in fish. The alterations in the lipid profiles observed after Ppar agonist exposure suggest that other chemicals with similar Ppar receptor affinities may cause disturbances in the lipid regulation of fish. Model organism: Atlantic cod (Gadus morhua). LSID: urn:lsid:zoobank.org:act:389BE401-2718-4CF2-BBAE-2E13A97A5E7B. COL Identifier: 6K72F.

GUBS: Graph-Based Unsupervised Brain Segmentation in MRI Images.

Brain segmentation in magnetic resonance imaging (MRI) images is the process of isolating the brain from non-brain tissues to simplify the further analysis, such as detecting pathology or calculating volumes. This paper proposes a Graph-based Unsupervised Brain Segmentation (GUBS) that processes 3D MRI images and segments them into brain, non-brain tissues, and backgrounds. GUBS first constructs an adjacency graph from a preprocessed MRI image, weights it by the difference between voxel intensities, and computes its minimum spanning tree (MST). It then uses domain knowledge about the different regions of MRIs to sample representative points from the brain, non-brain, and background regions of the MRI image. The adjacency graph nodes corresponding to sampled points in each region are identified and used as the terminal nodes for paths connecting the regions in the MST. GUBS then computes a subgraph of the MST by first removing the longest edge of the path connecting the terminal nodes in the brain and other regions, followed by removing the longest edge of the path connecting non-brain and background regions. This process results in three labeled, connected components, whose labels are used to segment the brain, non-brain tissues, and the background. GUBS was tested by segmenting 3D T1 weighted MRI images from three publicly available data sets. GUBS shows comparable results to the state-of-the-art methods in terms of performance. However, many competing methods rely on having labeled data available for training. Labeling is a time-intensive and costly process, and a big advantage of GUBS is that it does not require labels.

Epistemic uncertainty quantification in deep learning classification by the Delta method.

The Delta method is a classical procedure for quantifying epistemic uncertainty in statistical models, but its direct application to deep neural networks is prevented by the large number of parameters P. We propose a low cost approximation of the Delta method applicable to L2-regularized deep neural networks based on the top K eigenpairs of the Fisher information matrix. We address efficient computation of full-rank approximate eigendecompositions in terms of the exact inverse Hessian, the inverse outer-products of gradients approximation and the so-called Sandwich estimator. Moreover, we provide bounds on the approximation error for the uncertainty of the predictive class probabilities. We show that when the smallest computed eigenvalue of the Fisher information matrix is near the L2-regularization rate, the approximation error will be close to zero even when K≪P. A demonstration of the methodology is presented using a TensorFlow implementation, and we show that meaningful rankings of images based on predictive uncertainty can be obtained for two LeNet and ResNet-based neural networks using the MNIST and CIFAR-10 datasets. Further, we observe that false positives have on average a higher predictive epistemic uncertainty than true positives. This suggests that there is supplementing information in the uncertainty measure not captured by the classification alone.

Candidate Markers for Stratification and Classification in Rheumatoid Arthritis.

Rheumatoid arthritis (RA) is a chronic autoimmune, inflammatory disease, characterized by synovitis in small- and medium-sized joints and, if not treated early and efficiently, joint damage, and destruction. RA is a heterogeneous disease with a plethora of treatment options. The pro-inflammatory cytokine tumor necrosis factor (TNF) plays a central role in the pathogenesis of RA, and TNF inhibitors effectively repress inflammatory activity in RA. Currently, treatment decisions are primarily based on empirics and economic considerations. However, the considerable interpatient variability in response to treatment is a challenge. Markers for a more exact patient classification and stratification are lacking. The objective of this study was to identify markers in immune cell populations that distinguish RA patients from healthy donors with an emphasis on TNF signaling. We employed mass cytometry (CyTOF) with a panel of 13 phenotyping and 10 functional markers to explore signaling in unstimulated and TNF-stimulated peripheral blood mononuclear cells from 20 newly diagnosed, untreated RA patients and 20 healthy donors. The resulting high-dimensional data were analyzed in three independent analysis pipelines, characterized by differences in both data clean-up, identification of cell subsets/clustering and statistical approaches. All three analysis pipelines identified p-p38, IkBa, p-cJun, p-NFkB, and CD86 in cells of both the innate arm (myeloid dendritic cells and classical monocytes) and the adaptive arm (memory CD4+ T cells) of the immune system as markers for differentiation between RA patients and healthy donors. Inclusion of the markers p-Akt and CD120b resulted in the correct classification of 18 of 20 RA patients and 17 of 20 healthy donors in regression modeling based on a combined model of basal and TNF-induced signal. Expression patterns in a set of functional markers and specific immune cell subsets were distinct in RA patients compared to healthy individuals. These signatures may support studies of disease pathogenesis, provide candidate markers for response, and non-response to TNF inhibitor treatment, and aid the identification of future therapeutic targets.

RNA-Seq analysis of transcriptome responses in Atlantic cod (Gadus morhua) precision-cut liver slices exposed to benzo[a]pyrene and 17α-ethynylestradiol.

Polycyclic aromatic hydrocarbons such as benzo[a]pyrene (BaP) that activate the aryl hydrocarbon receptor (Ahr) pathway, and endocrine disruptors acting through the estrogen receptor pathway are among environmental pollutants of major concern. In this work, we exposed Atlantic cod (Gadus morhua) precision-cut liver slices (PCLS) to BaP (10 nM and 1000 nM), ethynylestradiol (EE2) (10 nM and 1000 nM), and equimolar mixtures of BaP and EE2 (10 nM and 1000 nM) for 48 h, and performed RNA-Seq based transcriptome mapping followed by systematic bioinformatics analyses. Our gene expression analysis showed that several genes were differentially expressed in response to BaP and EE2 treatments in PCLS. Strong up-regulation of genes coding for the cytochrome P450 1a (Cyp1a) enzyme and the Ahr repressor (Ahrrb) was observed in BaP treated PCLS. EE2 treatment of liver slices strongly up-regulated genes coding for precursors of vitellogenin (Vtg) and eggshell zona pellucida (Zp) proteins. As expected, pathway enrichment and network analysis showed that the Ahr and estrogen receptor pathways are among the top affected by BaP and EE2 treatments, respectively. Interestingly, two genes coding for fibroblast growth factor 3 (Fgf3) and fibroblast growth factor 4 (Fgf4) were up-regulated by EE2 in this study. To our knowledge, the fgf3 and fgf4 genes have not previously been described in relation to estrogen signaling in fish liver, and these results suggest the modulation of the FGF signaling pathway by estrogens in fish. The signature expression profiles of top differentially expressed genes in response to the single compound (BaP or EE2) treatment were generally maintained in the expression responses to the equimolar binary mixtures. However, in the mixture-treated groups, BaP appeared to have anti-estrogenic effects as observed by lower number of differentially expressed putative EE2 responsive genes. Our in-depth quantitative analysis of changes in liver transcriptome in response to BaP and EE2, using PCLS tissue culture provides further mechanistic insights into effects of the compounds. Moreover, the analyses demonstrate the usefulness of PCLS in cod for omics experiments.

The use of heuristics in intuitive mathematical judgment.

Anecdotal evidence points to the use of beauty as an indication of truth in mathematical problem solving. In the two experiments of the present study, we examined the use of heuristics and tested the assumption that participants use symmetry as a cue for correctness in an arithmetic verification task. We manipulated the symmetry of sets of dot pattern addition equations. Speeded decisions about the correctness of these equations led to higher endorsements for both correct and incorrect equations when the addend and sum dot patterns were symmetrical. Therefore, this effect is not due to the fact that symmetry facilitates calculation or estimation. We found systematic evidence for the use of heuristics in solving mathematical tasks, and we discuss how these findings relate to a processing-fluency account of intuition in mathematical judgment.