Identifying gene subnetworks associated with clinical outcome in ovarian cancer using Network Based Coalition Game

A Feature Selection Method Based on Shapley Value to False Alarm Reduction in ICUs A Genetic-Algorithm Approach

High false alarm rate in intensive care units (ICUs) has been identified as one of the most critical medical challenges in recent years. This often results in overwhelming the clinical staff by numerous false or unurgent alarms and decreasing the quality of care through enhancing the probability of missing true alarms as well as causing delirium, stress, sleep deprivation and depressed immune systems for patients. One major cause of false alarms in clinical practice is that the collected signals from different devices are processed individually to trigger an alarm, while there exists a considerable chance that the signal collected from one device is corrupted by noise or motion artifacts. In this paper, we propose a low-computational complexity yet accurate game-theoretic feature selection method which is based on a genetic algorithm that identifies the most informative biomarkers across the signals collected from various monitoring devices and can considerably reduce the rate of false alarms ¹.

Game Theoretic Approach for Systematic Feature Selection; Application in False Alarm Detection in Intensive Care Units

Intensive Care Units (ICUs) are equipped with many sophisticated sensors and monitoring devices to provide the highest quality of care for critically ill patients. However, these devices might generate false alarms that reduce standard of care and result in desensitization of caregivers to alarms. Therefore, reducing the number of false alarms is of great importance. Many approaches such as signal processing and machine learning, and designing more accurate sensors have been developed for this purpose. However, the significant intrinsic correlation among the extracted features from different sensors has been mostly overlooked. A majority of current data mining techniques fail to capture such correlation among the collected signals from different sensors that limits their alarm recognition capabilities. Here, we propose a novel information-theoretic predictive modeling technique based on the idea of coalition game theory to enhance the accuracy of false alarm detection in ICUs by accounting for the synergistic power of signal attributes in the feature selection stage. This approach brings together techniques from information theory and game theory to account for inter-features mutual information in determining the most correlated predictors with respect to false alarm by calculating Banzhaf power of each feature. The numerical results show that the proposed method can enhance classification accuracy and improve the area under the ROC (receiver operating characteristic) curve compared to other feature selection techniques, when integrated in classifiers such as Bayes-Net that consider inter-features dependencies.

Novel biomolecule lycopene-reduced graphene oxide-silver nanoparticle enhances apoptotic potential of trichostatin A in human ovarian cancer cells (SKOV3)

Background

Recently, there has been much interest in the field of nanomedicine to improve prevention, diagnosis, and treatment. Combination therapy seems to be most effective when two different molecules that work by different mechanisms are combined at low dose, thereby decreasing the possibility of drug resistance and occurrence of unbearable side effects. Based on this consideration, the study was designed to investigate the combination effect of reduced graphene oxide-silver nanoparticles (rGO-AgNPs) and trichostatin A (TSA) in human ovarian cancer cells (SKOV3).

Methods

The rGO-AgNPs were synthesized using a biomolecule called lycopene, and the resultant product was characterized by various analytical techniques. The combination effect of rGO-Ag and TSA was investigated in SKOV3 cells using various cellular assays such as cell viability, cytotoxicity, and immunofluorescence analysis.

Results

AgNPs were uniformly distributed on the surface of graphene sheet with an average size between 10 and 50 nm. rGO-Ag and TSA were found to inhibit cell viability in a dose-dependent manner. The combination of rGO-Ag and TSA at low concentration showed a significant effect on cell viability, and increased cytotoxicity by increasing the level of malondialdehyde and decreasing the level of glutathione, and also causing mitochondrial dysfunction. Furthermore, the combination of rGO-Ag and TSA had a more pronounced effect on DNA fragmentation and double-strand breaks, and eventually induced apoptosis.

Conclusion

This study is the first to report that the combination of rGO-Ag and TSA can cause potential cytotoxicity and also induce significantly greater cell death compared to either rGO-Ag alone or TSA alone in SKOV3 cells by various mechanisms including reactive oxygen species generation, mitochondrial dysfunction, and DNA damage. Therefore, this combination chemotherapy could be possibly used in advanced cancers that are not suitable for radiation therapy or surgical treatment and facilitate overcoming tumor resistance and disease progression.

Incorporating networks in a probabilistic graphical model to find drivers for complex human diseases

Discovering genetic mechanisms driving complex diseases is a hard problem. Existing methods often lack power to identify the set of responsible genes. Protein-protein interaction networks have been shown to boost power when detecting gene-disease associations. We introduce a Bayesian framework, Conflux, to find disease associated genes from exome sequencing data using networks as a prior. There are two main advantages to using networks within a probabilistic graphical model. First, networks are noisy and incomplete, a substantial impediment to gene discovery. Incorporating networks into the structure of a probabilistic models for gene inference has less impact on the solution than relying on the noisy network structure directly. Second, using a Bayesian framework we can keep track of the uncertainty of each gene being associated with the phenotype rather than returning a fixed list of genes. We first show that using networks clearly improves gene detection compared to individual gene testing. We then show consistently improved performance of Conflux compared to the state-of-the-art diffusion network-based method Hotnet2 and a variety of other network and variant aggregation methods, using randomly generated and literature-reported gene sets. We test Hotnet2 and Conflux on several network configurations to reveal biases and patterns of false positives and false negatives in each case. Our experiments show that our novel Bayesian framework Conflux incorporates many of the advantages of the current state-of-the-art methods, while offering more flexibility and improved power in many gene-disease association scenarios.

Networks and pathway-based methods are commonly used to improve the power of gene detection in associations with complex human diseases. Network diffusion approaches have shown their effectiveness and superior performance in cancer studies. Still, there are many problems such as noise and missingness with currently available human networks that bias the results of gene detection. We propose a novel graphical model-based method Conflux that overcomes several of the pitfalls of the existing state-of-the-art approaches while building on their successes. Conflux integrates genotype data with networks directly, using diffusion-like methods, but only as part of a structure in a probabilistic model to reduce the negative effect of the noise in the networks. This Bayesian framework allows Conflux to keep track of the uncertainty in the gene list that is being associated with the disease and consequently rank the genes with respect to our confidence in the association. It also allows for the discovery of gene sets that are not fully supported by the network if they have enough support in the data. These improvements result in a flexible approach that improves the power in many gene-disease association scenarios while reducing the number of false positives reported.

Network-Based Enriched Gene Subnetwork Identification: A Game-Theoretic Approach

Identifying subsets of genes that jointly mediate cancer etiology, progression, or therapy response remains a challenging problem due to the complexity and heterogeneity in cancer biology, a problem further exacerbated by the relatively small number of cancer samples profiled as compared with the sheer number of potential molecular factors involved. Pure data-driven methods that merely rely on multiomics data have been successful in discovering potentially functional genes but suffer from high false-positive rates and tend to report subsets of genes whose biological interrelationships are unclear. Recently, integrative data-driven models have been developed to integrate multiomics data with signaling pathway networks in order to identify pathways associated with clinical or biological phenotypes. However, these approaches suffer from an important drawback of being restricted to previously discovered pathway structures and miss novel genomic interactions as well as potential crosstalk among the pathways. In this article, we propose a novel coalition-based game-theoretic approach to overcome the challenge of identifying biologically relevant gene subnetworks associated with disease phenotypes. The algorithm starts from a set of seed genes and traverses a protein–protein interaction network to identify modulated subnetworks. The optimal set of modulated subnetworks is identified using Shapley value that accounts for both individual and collective utility of the subnetwork of genes. The algorithm is applied to two illustrative applications, including the identification of subnetworks associated with (i) disease progression risk in response to platinum-based therapy in ovarian cancer and (ii) immune infiltration in triple-negative breast cancer. The results demonstrate an improved predictive power of the proposed method when compared with state-of-the-art feature selection methods, with the added advantage of identifying novel potentially functional gene subnetworks that may provide insights into the mechanisms underlying cancer progression.