Emoke Ágnes Horvát

Global microRNA level regulation of EGFR-driven cell-cycle protein network in breast cancer

The EGFR‐driven cell‐cycle pathway has been extensively studied due to its pivotal role in breast cancer proliferation and pathogenesis. Although several studies reported regulation of individual pathway components by microRNAs (miRNAs), little is known about how miRNAs coordinate the EGFR protein network on a global miRNA (miRNome) level. Here, we combined a large‐scale miRNA screening approach with a high‐throughput proteomic readout and network‐based data analysis to identify which miRNAs are involved, and to uncover potential regulatory patterns. Our results indicated that the regulation of proteins by miRNAs is dominated by the nucleotide matching mechanism between seed sequences of the miRNAs and 3′‐UTR of target genes. Furthermore, the novel network‐analysis methodology we developed implied the existence of consistent intrinsic regulatory patterns where miRNAs simultaneously co‐regulate several proteins acting in the same functional module. Finally, our approach led us to identify and validate three miRNAs (miR‐124, miR‐147 and miR‐193a‐3p) as novel tumor suppressors that co‐target EGFR‐driven cell‐cycle network proteins and inhibit cell‐cycle progression and proliferation in breast cancer.

One plus one makes three (for social networks).

Members of social network platforms often choose to reveal private information, and thus sacrifice some of their privacy, in exchange for the manifold opportunities and amenities offered by such platforms. In this article, we show that the seemingly innocuous combination of knowledge of confirmed contacts between members on the one hand and their email contacts to non-members on the other hand provides enough information to deduce a substantial proportion of relationships between non-members. Using machine learning we achieve an area under the (receiver operating characteristic) curve () of at least for predicting whether two non-members known by the same member are connected or not, even for conservative estimates of the overall proportion of members, and the proportion of members disclosing their contacts.

A network-based method to assess the statistical significance of mild co-regulation effects.

Recent development of high-throughput, multiplexing technology has initiated projects that systematically investigate interactions between two types of components in biological networks, for instance transcription factors and promoter sequences, or microRNAs (miRNAs) and mRNAs. In terms of network biology, such screening approaches primarily attempt to elucidate relations between biological components of two distinct types, which can be represented as edges between nodes in a bipartite graph. However, it is often desirable not only to determine regulatory relationships between nodes of different types, but also to understand the connection patterns of nodes of the same type. Especially interesting is the co-occurrence of two nodes of the same type, i.e., the number of their common neighbours, which current high-throughput screening analysis fails to address. The co-occurrence gives the number of circumstances under which both of the biological components are influenced in the same way. Here we present SICORE, a novel network-based method to detect pairs of nodes with a statistically significant co-occurrence. We first show the stability of the proposed method on artificial data sets: when randomly adding and deleting observations we obtain reliable results even with noise exceeding the expected level in large-scale experiments. Subsequently, we illustrate the viability of the method based on the analysis of a proteomic screening data set to reveal regulatory patterns of human microRNAs targeting proteins in the EGFR-driven cell cycle signalling system. Since statistically significant co-occurrence may indicate functional synergy and the mechanisms underlying canalization, and thus hold promise in drug target identification and therapeutic development, we provide a platform-independent implementation of SICORE with a graphical user interface as a novel tool in the arsenal of high-throughput screening analysis.

Measuring long-term impact based on network centrality: unraveling cinematic citations.

Traditional measures of success for film, such as box-office revenue and critical acclaim, lack the ability to quantify long-lasting impact and depend on factors that are largely external to the craft itself. With the growing number of films that are being created and large-scale data becoming available through crowd-sourced online platforms, an endogenous measure of success that is not reliant on manual appraisal is of increasing importance. In this article we propose such a ranking method based on a combination of centrality indices. We apply the method to a network that contains several types of citations between more than 40,000 international feature films. From this network we derive a list of milestone films, which can be considered to constitute the foundations of cinema. In a comparison to various existing lists of ‘greatest’ films, such as personal favourite lists, voting lists, lists of individual experts, and lists deduced from expert polls, the selection of milestone films is more diverse in terms of genres, actors, and main creators. Our results shed light on the potential of a systematic quantitative investigation based on cinematic influences in identifying the most inspiring creations in world cinema. In a broader perspective, we introduce a novel research question to large-scale citation analysis, one of the most intriguing topics that have been at the forefront of scientific enquiries for the past fifty years and have led to the development of various network analytic methods. In doing so, we transfer widely studied approaches from citation analysis to the the newly emerging field of quantification efforts in the arts. The specific contribution of this paper consists in modelling the multidimensional cinematic references as a growing multiplex network and in developing a methodology for the identification of central films in this network.

Different flavors of randomness: comparing random graph models with fixed degree sequences

When a structural characteristic of a network is measured, the observed value needs to be compared to its expected value in a random graph model to assess the statistical significance of its occurrence. The random graph model with which the observed graph is compared is chosen to be structurally similar to the real-world network in some aspects and totally random in all others. To make the analysis of the expected value amenable, the random graph model is also chosen to be as simple as possible. The most common random graph models maintain the degree sequence of the observed graph or at least approximate it. In cases where multi-edges and self-loops are not allowed, typically the fixed degree sequence model (FDSM) is used. Since it is computationally expensive, in this article, we discuss whether one of the following three approximative models can replace it: the configuration model, its simplified version (eCFG), and the mathematical approximation we term simple independence model. While the latter models are more scalable than the FDSM, we show that there are several networks for which they cannot be meaningfully applied. We investigate based on some examples whether, and if so in which cases, these approximating models can replace the computationally more involved FDSM.

Assessing Low-Intensity Relationships in Complex Networks.

Many large network data sets are noisy and contain links representing low-intensity relationships that are difficult to differentiate from random interactions. This is especially relevant for high-throughput data from systems biology, large-scale ecological data, but also for Web 2.0 data on human interactions. In these networks with missing and spurious links, it is possible to refine the data based on the principle of structural similarity, which assesses the shared neighborhood of two nodes. By using similarity measures to globally rank all possible links and choosing the top-ranked pairs, true links can be validated, missing links inferred, and spurious observations removed. While many similarity measures have been proposed to this end, there is no general consensus on which one to use. In this article, we first contribute a set of benchmarks for complex networks from three different settings (e-commerce, systems biology, and social networks) and thus enable a quantitative performance analysis of classic node similarity measures. Based on this, we then propose a new methodology for link assessment called z* that assesses the statistical significance of the number of their common neighbors by comparison with the expected value in a suitably chosen random graph model and which is a consistently top-performing algorithm for all benchmarks. In addition to a global ranking of links, we also use this method to identify the most similar neighbors of each single node in a local ranking, thereby showing the versatility of the method in two distinct scenarios and augmenting its applicability. Finally, we perform an exploratory analysis on an oceanographic plankton data set and find that the distribution of microbes follows similar biogeographic rules as those of macroorganisms, a result that rejects the global dispersal hypothesis for microbes.

Network vs Market Relations: The Effect of Friends in Crowdfunding

Crowds offer a new form of efficacious collective decision making, yet knowledge about the mechanisms by which they achieve superior outcomes remains nascent. It has been suggested that crowds work best with market-like relationships when individuals make independent decisions and possess dissimilar information. By contrast, sociological discussions of markets argue that risky decisions are mitigated by network relations that embed economic transactions in social ties that promote trustworthiness and reciprocity. To investigate the role of networks within crowds and their performance effects, we examined the complete record of financial lending decisions on Prosper.com, 1/2006-3/2012, the first U.S. crowdfunding platform and a chief gateway to capital for entrepreneurs and general borrowers that continues to disrupt conventional financial lending structures infusing more than

SICOP: identifying significant co-interaction patterns

5.1 billion into the market in 2013. Our study reveals how reciprocity, recurring borrower-lender dyads, and persistent co-lending underpin the dynamics of network lending. Further, we show how network ties influence the evolution of the lending behavior. We find that in the early stage of fundraising, network relations provide larger proportions of loans, typically lending four times more per bid than strangers. They also respond to loan requests on average 59.5% sooner than strangers. The size of the first loan and the time to lending also tend to prompt lending by strangers, suggesting that network relations might move the market, a finding that persists even as fewer lenders dominate more of the market for loans on Prosper. Finally, network relations are associated with greater engagement: when the first loan is underwritten by a friend, 50% of the remaining loans come from friends as well.

Spring-block approach for crack patterns in glass

SUMMARY Interactions between various types of molecules that regulate crucial cellular processes are extensively investigated by high-throughput experiments and require dedicated computational methods for the analysis of the resulting data. In many cases, these data can be represented as a bipartite graph because it describes interactions between elements of two different types such as the influence of different experimental conditions on cellular variables or the direct interaction between receptors and their activators/inhibitors. One of the major challenges in the analysis of such noisy datasets is the statistical evaluation of the relationship between any two elements of the same type. Here, we present SICOP (significant co-interaction patterns), an implementation of a method that provides such an evaluation based on the number of their common interaction partners, their so-called co-interaction. This general network analytic method, proved successful in diverse fields, provides a framework for assessing the significance of this relationship by comparison with the expected co-interaction in a suitable null model of the same bipartite graph. SICOP takes into consideration up to two distinct types of interactions such as up- or downregulation. The tool is written in Java and accepts several common input formats and supports different output formats, facilitating further analysis and visualization. Its key features include a user-friendly interface, easy installation and platform independence. AVAILABILITY The software is open source and available at cna.cs.uni-kl.de/SICOP under the terms of the GNU General Public Licence (version 3 or later).

Peer-to-peer lending and bias in crowd decision-making

Fracture patterns resulting from point-like impact acting perpendicularly on the plane of a commercial sodalime glass plate is modelled by a spring-block system. The characteristic patterns consist of crack lines that are spreading radially from the impact point and concentric arcs intersecting these radial lines. Experimental results suggest that the number of radial crack lines is scaling linearly with the energy dissipated during the crack formation process. The elaborated spring-block model reproduces with success the observed fracture patterns and scaling law.