Explore the vast range of titles available.

Background The selection and prioritization of drug targets is a central problem in drug discovery. Computational approaches can leverage the growing number of large-scale human genomics and proteomics data to make in-silico target identification, reducing the cost and the time needed. Results We developed a machine learning approach to score proteins to generate a druggability score of novel targets. In our model we incorporated 70 protein features which included properties derived from the sequence, features characterizing protein functions as well as network properties derived from the protein-protein interaction network. The advantage of this approach is that it is unbiased and even less studied proteins with limited information about their function can score well as most of the features are independent of the accumulated literature. We build models on a training set which consist of targets with approved drugs and a negative set of non-drug targets. The machine learning techniques help to identify the most important combination of features differentiating validated targets from non-targets. We validated our predictions on an independent set of clinical trial drug targets, achieving a high accuracy characterized by an Area Under the Curve (AUC) of 0.89. Our most predictive features included biological function of proteins, network centrality measures, protein essentiality, tissue specificity, localization and solvent accessibility. Our predictions, based on a small set of 102 validated oncology targets, recovered the majority of known drug targets and identifies a novel set of proteins as drug target candidates. Conclusions We developed a machine learning approach to prioritize proteins according to their similarity to approved drug targets. We have shown that the method proposed is highly predictive on a validation dataset consisting of 277 targets of clinical trial drug confirming that our computational approach is an efficient and cost-effective tool for drug target discovery and prioritization. Our predictions were based on oncology targets and cancer relevant biological functions, resulting in significantly higher scores for targets of oncology clinical trial drugs compared to the scores of targets of trial drugs for other indications. Our approach can be used to make indication specific drug-target prediction by combining generic druggability features with indication specific biological functions.

The identification of human proteins that are amenable to pharmacologic modulation without significant off-target effects remains an important unsolved challenge. Computational methods have been devised to identify features which distinguish between “druggable” and “undruggable” proteins, finding that protein sequence, tissue and cellular localization, biological role, and position in the protein–protein interaction network are all important discriminant factors. However, many prior efforts to automate the assessment of protein druggability suffer from low performance or poor interpretability. We developed a neural network-based machine learning model capable of generating druggability sub-scores based on each of four distinct categories, combining them to form an overall druggability score. The model achieves an excellent performance in separating drugged and undrugged proteins in the human proteome, with an area under the receiver operating characteristic (AUC) of 0.95. Our use of multiple sub-scores allows the assessment of potential protein targets of interest based on distinct contributors to druggability, leading to a more interpretable and holistic model to identify novel targets.

Eribulin mesylate is a synthetic macrocyclic ketone analog of the marine sponge natural product halichondrin B. Eribulin is a mechanistically unique inhibitor of microtubule dynamics. In this study, we investigated whether selective signal pathways were associated with eribulin activity compared to paclitaxel, which stabilizes microtubules, based on gene expression profiling of cell line panels of breast, endometrial, and ovarian cancer in vitro. We determined the sets of genes that were differentially altered between eribulin and paclitaxel treatment in breast, endometrial, and ovarian cancer cell line panels. Our unsupervised clustering analyses revealed that expression profiles of gene sets altered with treatments were correlated with the in vitro antiproliferative activities of the drugs. Several tubulin isotypes had significantly lower expression in cell lines treated with eribulin compared to paclitaxel. Pathway enrichment analyses of gene sets revealed that the common pathways altered between treatments in the 3 cancer panels were related to cytoskeleton remodeling and cell cycle regulation. The epithelial-mesenchymal transition (EMT) pathway was enriched in genes with significantly altered expression between the two drugs for breast and endometrial cancers, but not for ovarian cancer. Expression of genes from the EMT pathway correlated with eribulin sensitivity in breast cancer and with paclitaxel sensitivity in endometrial cancer. Alteration of expression profiles of EMT genes between sensitive and resistant cell lines allowed us to predict drug sensitivity for breast and endometrial cancers. Gene expression analysis showed that gene sets that were altered between eribulin and paclitaxel correlated with drug in vitro antiproliferative activities in breast and endometrial cancer cell line panels. Among the panels, breast cancer provided the strongest differentiation between eribulin and paclitaxel sensitivities based on gene expression. In addition, EMT genes were predictive of eribulin sensitivity in the breast and endometrial cancer panels.

Background The identification of key target nodes within complex molecular networks remains a common objective in scientific research. The results of pathway analyses are usually sets of fairly complex networks or functional processes that are deemed relevant to the condition represented by the molecular profile. To be useful in a research or clinical laboratory, the results need to be translated to the level of testable hypotheses about individual genes and proteins within the condition of interest. Results In this paper we describe novel computational methodology capable of predicting key regulatory genes and proteins in disease- and condition-specific biological networks. The algorithm builds shortest path network connecting condition-specific genes (e.g. differentially expressed genes) using global database of protein interactions from MetaCore. We evaluate the number of all paths traversing each node in the shortest path network in relation to the total number of paths going via the same node in the global network. Using these numbers and the relative size of the initial data set, we determine the statistical significance of the network connectivity provided through each node. We applied this method to gene expression data from psoriasis patients and identified many confirmed biological targets of psoriasis and suggested several new targets. Using predicted regulatory nodes we were able to reconstruct disease pathways that are in excellent agreement with the current knowledge on the pathogenesis of psoriasis. Conclusion The systematic and automated approach described in this paper is readily applicable to uncovering high-quality therapeutic targets, and holds great promise for developing network-based combinational treatment strategies for a wide range of diseases.

The problem of prostate cancer progression to androgen independence has been extensively studied. Several studies systematically analyzed gene expression profiles in the context of biological networks and pathways, uncovering novel aspects of prostate cancer. Despite significant research efforts, the mechanisms underlying tumor progression are poorly understood. We applied a novel approach to reconstruct system-wide molecular events following stimulation of LNCaP prostate cancer cells with synthetic androgen and to identify potential mechanisms of androgen-independent progression of prostate cancer. We have performed concurrent measurements of gene expression and protein levels following the treatment using microarrays and iTRAQ proteomics. Sets of up-regulated genes and proteins were analyzed using our novel concept of \"topological significance\". This method combines high-throughput molecular data with the global network of protein interactions to identify nodes which occupy significant network positions with respect to differentially expressed genes or proteins. Our analysis identified the network of growth factor regulation of cell cycle as the main response module for androgen treatment in LNCap cells. We show that the majority of signaling nodes in this network occupy significant positions with respect to the observed gene expression and proteomic profiles elicited by androgen stimulus. Our results further indicate that growth factor signaling probably represents a \"second phase\" response, not directly dependent on the initial androgen stimulus. We conclude that in prostate cancer cells the proliferative signals are likely to be transmitted from multiple growth factor receptors by a multitude of signaling pathways converging on several key regulators of cell proliferation such as c-Myc, Cyclin D and CREB1. Moreover, these pathways are not isolated but constitute an interconnected network module containing many alternative routes from inputs to outputs. If the whole network is involved, a precisely formulated combination therapy may be required to fight the tumor growth effectively.

Human cancer is caused by the accumulation of mutations in oncogenes and tumor suppressor genes. To catalog the genetic changes that occur during tumorigenesis, we isolated DNA from 11 breast and 11 colorectal tumors and determined the sequences of the genes in the Reference Sequence database in these samples. Based on analysis of exons representing 20,857 transcripts from 18,191 genes, we conclude that the genomic landscapes of breast and colorectal cancers are composed of a handful of commonly mutated gene \"mountains\" and a much larger number of gene \"hills\" that are mutated at low frequency. We describe statistical and bioinformatic tools that may help identify mutations with a role in tumorigenesis. These results have implications for understanding the nature and heterogeneity of human cancers and for using personal genomics for tumor diagnosis and therapy.

The selection and prioritization of drug targets is a central problem in drug discovery. Computational approaches can leverage the growing number of large-scale human genomics and proteomics data to make in-silico target identification, reducing the cost and the time needed. We developed a machine learning approach to score proteins to generate a druggability score of novel targets. In our model we incorporated 70 protein features which included properties derived from the sequence, features characterizing protein functions as well as network properties derived from the protein-protein interaction network. The advantage of this approach is that it is unbiased and even less studied proteins with limited information about their function can score well as most of the features are independent of the accumulated literature. We build models on a training set which consist of targets with approved drugs and a negative set of non-drug targets. The machine learning techniques help to identify the most important combination of features differentiating validated targets from non-targets. We validated our predictions on an independent set of clinical trial drug targets, achieving a high accuracy characterized by an Area Under the Curve (AUC) of 0.89. Our most predictive features included biological function of proteins, network centrality measures, protein essentiality, tissue specificity, localization and solvent accessibility. Our predictions, based on a small set of 102 validated oncology targets, recovered the majority of known drug targets and identifies a novel set of proteins as drug target candidates. We developed a machine learning approach to prioritize proteins according to their similarity to approved drug targets. We have shown that the method proposed is highly predictive on a validation dataset consisting of 277 targets of clinical trial drug confirming that our computational approach is an efficient and cost-effective tool for drug target discovery and prioritization. Our predictions were based on oncology targets and cancer relevant biological functions, resulting in significantly higher scores for targets of oncology clinical trial drugs compared to the scores of targets of trial drugs for other indications. Our approach can be used to make indication specific drug-target prediction by combining generic druggability features with indication specific biological functions.

A minimally invasive diagnostic assay for early detection of Alzheimer's disease (AD) is required to select optimal patient groups in clinical trials, monitor disease progression and response to treatment, and to better plan patient clinical care. Blood is an attractive source for biomarkers due to minimal discomfort to the patient, encouraging greater compliance in clinical trials and frequent testing. MiRNAs belong to the class of non-coding regulatory RNA molecules of ∼22 nt length and are now recognized to regulate ∼60% of all known genes through post-transcriptional gene silencing (RNAi). They have potential as useful biomarkers for clinical use because of their stability and ease of detection in many tissues, especially blood. Circulating profiles of miRNAs have been shown to discriminate different tumor types, indicate staging and progression of the disease and to be useful as prognostic markers. Recently their role in neurodegenerative diseases, both as diagnostic biomarkers as well as explaining basic disease etiology has come into focus. Here we report the discovery and validation of a unique circulating 7-miRNA signature (hsa-let-7d-5p, hsa-let-7g-5p, hsa-miR-15b-5p, hsa-miR-142-3p, hsa-miR-191-5p, hsa-miR-301a-3p and hsa-miR-545-3p) in plasma, which could distinguish AD patients from normal controls (NC) with >95% accuracy (AUC of 0.953). There was a >2 fold difference for all signature miRNAs between the AD and NC samples, with p-values<0.05. Pathway analysis, taking into account enriched target mRNAs for these signature miRNAs was also carried out, suggesting that the disturbance of multiple enzymatic pathways including lipid metabolism could play a role in AD etiology.

Cancer dependency maps have accelerated the discovery of essential genes and potential drug targets. Here we used machine learning to build translational dependency maps of patients’ tumors and normal tissue biopsies, which identified oncogenes and synthetic lethalities that are predictive of drug responses and patients’ outcomes.

Background Astrocyte activation is a characteristic response to injury in the central nervous system, and can be either neurotoxic or neuroprotective, while the regulation of both roles remains elusive. Methods To decipher the regulatory elements controlling astrocyte-mediated neurotoxicity in glaucoma, we conducted a systems-level functional analysis of gene expression, proteomic and genetic data associated with reactive optic nerve head astrocytes (ONHAs). Results Our reconstruction of the molecular interactions affected by glaucoma revealed multi-domain biological networks controlling activation of ONHAs at the level of intercellular stimuli, intracellular signaling and core effectors. The analysis revealed that synergistic action of the transcription factors AP-1, vitamin D receptor and Nuclear Factor-kappaB in cross-activation of multiple pathways, including inflammatory cytokines, complement, clusterin, ephrins, and multiple metabolic pathways. We found that the products of over two thirds of genes linked to glaucoma by genetic analysis can be functionally interconnected into one epistatic network via experimentally-validated interactions. Finally, we built and analyzed an integrative disease pathology network from a combined set of genes revealed in genetic studies, genes differentially expressed in glaucoma and closely connected genes/proteins in the interactome. Conclusion Our results suggest several key biological network modules that are involved in regulating neurotoxicity of reactive astrocytes in glaucoma, and comprise potential targets for cell-based therapy.