arXiv subjects

Steven J. M. Jones

Publications and source records attributed to Steven J. M. Jones.

4 recordsLinked to original sources

arXiv2020

Classification of Epithelial Ovarian Carcinoma Whole-Slide Pathology Images Using Deep Transfer Learning

Ovarian cancer is the most lethal cancer of the female reproductive organs. There are $5$ major histological subtypes of epithelial ovarian cancer, each with distinct morphological, genetic, and clinical features. Currently, these histotypes are determined by a pathologist's microscopic examination of tumor whole-slide images (WSI). This process has been hampered by poor inter-observer agreement (Cohen's kappa $0.54$-$0.67$). We utilized a \textit{two}-stage deep transfer learning algorithm based on convolutional neural networks (CNN) and progressive resizing for automatic classification of epithelial ovarian carcinoma WSIs. The proposed algorithm achieved a mean accuracy of $87.54\%$ and Cohen's kappa of $0.8106$ in the slide-level classification of $305$ WSIs; performing better than a standard CNN and pathologists without gynecology-specific training.

eess.IV↗

arXiv2015

New class of compounds - variators - are reprogramming substrate specificity of H4K12Ac, H4K16Ac and H4K20Ac epigenetic marks reading bromodomain of BPTF protein

Previously reported [http://arxiv.org/abs/1506.06433] reprogramming of substrate specificity of H3K4Me3 epigenetic marks reading PHD domain of BPTF protein illustrates therapeutic potential of a new class of non-inhibitor small organic compounds - variators. Here we address the question about reproducibility of rational design of variators by reprogramming of the second epigenetic marks reading domain of BPTF protein - bromodomain. Bromodomain of BPTF binds to epigenetic marks in form of acetylated lysine of histone H4 (H4K12Ac, H4K16Ac and H4K20Ac), which physicochemical properties and binding mode differs considerably from those of methylated H3K4 marks. Thus, detailed description of computational approach for reprogramming of bromodomain substrate specificity illustrates both general and target specific attributes of computer aided variators design.

q-bio.BM↗

arXiv2015

New class of compounds - variators - are reprogramming substrate specificity of H3K4me3 epigenetic marks reading PHD domain of BPTF protein

In lymphoma, mutations in genes of histone modifying proteins are frequently observed. Notably, somatic mutations in the activatory histone modification writing protein MLL2 and the repressive modification writer EZH2 are the most frequent. Gain of function mutations are typically detected in EZH2 whilst MLL2 mutations are usually observed as conferring a homozygous loss of function. The gain-of-function mutations in EZH2 provide an obvious target for the development of inhibitors with therapeutic potential. To counter the loss of functional MLL2 protein, we computationally predicted compounds that are able to modulate the reader of the corresponding modifications, BPTF, to recognize other forms of the histone H3 lysine 4, instead of the tri-methylated form normally produced by MLL2. By forming a synthetic triple-complex of a compound, the histone H3 tail and BPTF we potentially circumvent the requirement for functional MLL2 methyl-transferase through the modulation of BPTF activity. Here we show a proof-of-principle that special compounds, named variators, can reprogram selectivity of protein binding and thus create artificial regulatory pathways which can have a potential therapeutic role. A therapeutic role of BPTF variators may extend to other diseases that involve loss of MLL2 function, such as Kabuki syndrome or the aberrant functioning of H3K4 modification as observed in Huntington disease and in memory formation.

q-bio.BM↗

arXiv2009

KiWi: A Scalable Subspace Clustering Algorithm for Gene Expression Analysis

Subspace clustering has gained increasing popularity in the analysis of gene expression data. Among subspace cluster models, the recently introduced order-preserving sub-matrix (OPSM) has demonstrated high promise. An OPSM, essentially a pattern-based subspace cluster, is a subset of rows and columns in a data matrix for which all the rows induce the same linear ordering of columns. Existing OPSM discovery methods do not scale well to increasingly large expression datasets. In particular, twig clusters having few genes and many experiments incur explosive computational costs and are completely pruned off by existing methods. However, it is of particular interest to determine small groups of genes that are tightly coregulated across many conditions. In this paper, we present KiWi, an OPSM subspace clustering algorithm that is scalable to massive datasets, capable of discovering twig clusters and identifying negative as well as positive correlations. We extensively validate KiWi using relevant biological datasets and show that KiWi correctly assigns redundant probes to the same cluster, groups experiments with common clinical annotations, differentiates real promoter sequences from negative control sequences, and shows good association with cis-regulatory motif predictions.

cs.DB↗