Biology subjects

Wagner, B.

Publications and source records attributed to Wagner, B..

26 records · Page 2Linked to original sources

bioRxiv2019

BrainNET: Inference of brain network topology using Machine Learning

ObjectiveTo develop a new fMRI network inference method, BrainNET, that utilizes an efficient machine learning algorithm to quantify contributions of various regions of interests (ROIs) in the brain to a specific ROI. MethodsBrainNET is based on Extremely Randomized Trees (ERT) to estimate network topology from fMRI data and modified to generate an adjacency matrix representing brain network topology, without reliance on arbitrary thresholds. Open source simulated fMRI data of fifty subjects in twenty-eight different simulations under various confounding conditions with known ground truth was used to validate the method. Performance was compared with correlation and partial correlation (PC). The real-world performance was then evaluated in a publicly available Attention-deficit/hyperactivity disorder (ADHD) dataset including 134 Typically Developing Children (mean age: 12.03, males: 83), 75 ADHD Inattentive (mean age: 11.46, males: 56) and 93 ADHD Combined (mean age: 11.86, males: 77) subjects. Network topologies in ADHD were inferred using BrainNET, correlation, and PC. Graph metrics were extracted to determine differences between the ADHD groups. ResultsBrainNET demonstrated excellent performance across all simulations and varying confounders in identifying true presence of connections. In the ADHD dataset, BrainNET was able to identify significant changes (p< 0.05) in graph metrics between groups. No significant changes in graph metrics between ADHD groups was identified using correlation and PC.

bioengineering↗

bioRxiv2019

Multidimensional and Multiresolution Ensemble Networks for Brain Tumor Segmentation

In this work, we developed multiple 2D and 3D segmentation models with multiresolution input to segment brain tumor components, and then ensembled them to obtain robust segmentation maps. This reduced overfitting and resulted in a more generalized model. Multiparametric MR images of 335 subjects from BRATS 2019 challenge were used for training the models. Further, we tested a classical machine learning algorithm (xgboost) with features extracted from the segmentation maps to classify subject survival range. Preliminary results on the BRATS 2019 validation dataset demonstrasted this method can achieve excellent performance with DICE scores of 0.898, 0.784, 0.779 for whole tumor, tumor core and enhancing tumor respectively and accuracy 34.5 % for survuval prediction.

neuroscience↗

bioRxiv2019

Fully Automated Brain Tumor Segmentation and Survival Prediction of Gliomas using Deep Learning and MRI

Tumor segmentation of magnetic resonance (MR) images is a critical step in providing objective measures of predicting aggressiveness and response to therapy in gliomas. It has valuable applications in diagnosis, monitoring, and treatment planning of brain tumors. The purpose of this work was to develop a fully automated deep learning method for brain tumor segmentation and survival prediction. Well curated brain tumor cases with multi-parametric MR Images from the BraTS2019 dataset were used. A three-group framework was implemented, with each group consisting of three 3D-Dense-UNets to segment whole tumor (WT), tumor core (TC) and enhancing tumor (ET). This method was implemented to decompose the complex multi-class segmentation problem into individual binary segmentation problems for each sub-component. Each group was trained using different approaches and loss functions. The output segmentations of a particular label from their respective networks from the 3 groups were ensembled and post-processed. For survival analysis, a linear regression model based on imaging texture features and wavelet texture features extracted from each of the segmented components was implemented. The networks were tested on the BraTS2019 validation dataset including 125 cases for the brain tumor segmentation task and 29 cases for the survival prediction task. The segmentation networks achieved average dice scores of 0.901, 0.844 and 0.801 for WT, TC and ET respectively. The survival prediction network achieved an accuracy score of 0.55 and mean squared error (MSE) of 119244. This method could be implemented as a robust tool to assist clinicians in primary brain tumor management and follow-up.

neuroscience↗

bioRxiv2019

Classification of Brain Tumor IDH Status using MRI and Deep Learning

Isocitrate dehydrogenase (IDH) mutation status is an important marker in glioma diagnosis and therapy. We propose a novel automated pipeline for predicting IDH status noninvasively using deep learning and T2-weighted (T2w) MR images with minimal preprocessing (N4 bias correction and normalization to zero mean and unit variance). T2w MRI and genomic data were obtained from The Cancer Imaging Archive dataset (TCIA) for 260 subjects (120 High grade and 140 Low grade gliomas). A fully automated 2D densely connected model was trained to classify IDH mutation status on 208 subjects and tested on another held-out set of 52 subjects, using 5-fold cross validation. Data leakage was avoided by ensuring subject separation during the slice-wise randomization. Mean classification accuracy of 90.5% was achieved for each axial slice in predicting the three classes of no tumor, IDH mutated and IDH wild-type. Test accuracy of 83.8% was achieved in predicting IDH mutation status for individual subjects on the test dataset of 52 subjects. We demonstrate a deep learning method to predict IDH mutation status using T2w MRI alone. Radiologic imaging studies using deep learning methods must address data leakage (subject duplication) in the randomization process to avoid upward bias in the reported classification accuracy.

neuroscience↗

bioRxiv2019

The expression of equine keratins K42 and K124 is restricted to the hoof epidermal lamellae of Equus caballus

The equine hoof inner epithelium is folded into primary and secondary epidermal lamellae which increase the dermo-epidermal junction surface area of the hoof and can be affected by laminitis, a common disease of equids. Two keratin proteins (K), K42 and K124, are the most abundant keratins in the hoof lamellar tissue of Equus caballus. We hypothesize that these keratins are lamellar tissue-specific and could serve as differentiation- and disease-specific markers. Our objective was to characterize the expression of K42 and K124 in equine stratified epithelia and to generate monoclonal antibodies against K42 and K124. By RT-PCR analysis, keratin gene (KRT) KRT42 and KRT124 expression was present in lamellar tissue, but not cornea, haired skin, or hoof coronet. In situ hybridization studies showed that KRT124 localized to the suprabasal and, to a lesser extent, basal cells of the lamellae, was absent from haired skin and hoof coronet, and abruptly transitions from KRT124-negative coronet to KRT124-positive proximal lamellae. A monoclonal antibody generated against full-length recombinant equine K42 detected a lamellar keratin of the appropriate size, but also cross-reacted with other epidermal keratins. Three monoclonal antibodies generated against N- and C-terminal K124 peptides detected a band of the appropriate size in lamellar tissue and did not cross-react with proteins from haired skin, corneal limbus, hoof coronet, tongue, glabrous skin, oral mucosa, or chestnut on immunoblots. K124 localized to lamellar cells by indirect immunofluorescence. This is the first study to demonstrate the localization and expression of a hoof lamellar-specific keratin, K124, and to validate anti-K124 monoclonal antibodies.

cell biology↗

bioRxiv2019

Deep learning enables therapeutic antibody optimization in mammalian cells

Therapeutic antibody optimization is time and resource intensive, largely because it requires low-throughput screening (103 variants) of full-length IgG in mammalian cells, typically resulting in only a few optimized leads. Here, we use deep learning to interrogate and predict antigen-specificity from a massively diverse sequence space to identify globally optimized antibody variants. Using a mammalian display platform and the therapeutic antibody trastuzumab, rationally designed site-directed mutagenesis libraries are introduced by CRISPR/Cas9-mediated homology-directed repair (HDR). Screening and deep sequencing of relatively small libraries (104) produced high quality data capable of training deep neural networks that accurately predict antigen-binding based on antibody sequence. Deep learning is then used to predict millions of antigen binders from an in silico library of ~108 variants, where experimental testing of 30 randomly selected variants showed all 30 retained antigen specificity. The full set of in silico predicted binders is then subjected to multiple developability filters, resulting in thousands of highly-optimized lead candidates. With its scalability and capacity to interrogate high-dimensional protein sequence space, deep learning offers great potential for antibody engineering and optimization.

synthetic biology↗

bioRxiv2018

6-Phosphogluconate Dehydrogenase Links Cytosolic Carbohydrate Metabolism to Protein Secretion

The proteinaceous extracellular matrix (ECM) is vital for cancer cell survival, proliferation, migration, and differentiation. However, little is known regarding metabolic pathways required in the ECM secretion process. By using an unbiased computational approach, we searched for enzymes whose suppression may lead to disruptions in protein secretion. Here, we show that 6-phosphogluconate dehydrogenase (PGD), a cytosolic enzyme involved in carbohydrate metabolism, is required for endoplasmic reticulum (ER) structural integrity and protein secretion. Chemical inhibition or genetic suppression of its activity led to cell stress accompanied by significantly expanded ER volume and can be rescued by compensating glutathione supplies. Our results also suggest that this characteristic ER-dilation phenotype may be a general marker indicating increased ECM protein congestion inside cells and decreased secretion. Thus, PGD exemplifies a nexus of cytosolic carbohydrate metabolism and protein secretion.

biochemistry↗

bioRxiv2018

Engineering of new-to-nature halogenated indigo precursors in plants

Plants are versatile chemists producing a tremendous variety of specialized compounds. Here, we describe the engineering of entirely novel metabolic pathways in planta enabling generation of halogenated indigo precursors as non-natural plant products.\n\nIndican (indolyl-{beta}-D-glucopyranoside) is a secondary metabolite characteristic of a number of dyers plants. Its deglucosylation and subsequent oxidative dimerization leads to the blue dye, indigo. Halogenated indican derivatives are commonly used as detection reagents in histochemical and molecular biology applications; their production, however, relies largely on chemical synthesis. To attain the de novo biosynthesis in a plant-based system devoid of indican, we employed a sequence of enzymes from diverse sources, including three microbial tryptophan halogenases substituting the amino acid at either C5, C6, or C7 of the indole moiety. Subsequent processing of the halotryptophan by bacterial tryptophanase TnaA in concert with a mutant of the human cytochrome P450 monooxygenase 2A6 and glycosylation of the resulting indoxyl derivatives by an endogenous tobacco glucosyltransferase yielded corresponding haloindican variants in transiently transformed Nicotiana benthamiana plants. Accumulation levels were highest when the 5-halogenase PyrH was utilized, reaching 0.93 {+/-}0.089 mg/g dry weight of 5-chloroindican. The identity of the latter was unambiguously confirmed by NMR analysis. Moreover, our combinatorial approach, facilitated by the modular assembly capabilities of the GoldenBraid cloning system and inspired by the unique compartmentation of plant cells, afforded testing a number of alternative subcellular localizations for pathway design. In consequence, chloroplasts were validated as functional biosynthetic venues for haloindican, with the requisite reducing augmentation of the halogenases as well as the cytochrome P450 monooxygenase fulfilled by catalytic systems native to the organelle.\n\nThus, our study puts forward a viable alternative production platform for halogenated fine chemicals, eschewing reliance on fossil fuel resources and toxic chemicals. We further contend that in planta generation of halogenated indigoid precursors previously unknown to nature offers an extended view on and, indeed, pushes forward the established frontiers of biosynthetic capacity of plants.\n\nGraphical abstract\n\nHighlightsO_LIEntirely novel combinatorial pathway yielded new-to-nature halogenated indigoids.\nC_LIO_LIAn array of specific chloro- and bromoindican regioisomers was generated.\nC_LIO_LISignificant retrieval rates afforded unambiguous metabolite identification.\nC_LIO_LIEngineered plants emerge as alternative manufacture chassis for fine chemicals.\nC_LI

synthetic biology↗