Mining Large Datasets Research Articles

Advances in Cheminformatics Methodologies and Infrastructure to Support the Data Mining of Large, Heterogeneous Chemical Datasets

In recent years, there has been an explosion in the availability of publicly accessible chemical information, including chemical structures of small molecules, structure-derived properties and associated biological activities in a variety of assays. These data sources present us with a significant opportunity to develop and apply computational tools to extract and understand the underlying structure-activity relationships. Furthermore, by integrating chemical data sources with biological information (protein structure, gene expression and so on), we can attempt to build up a holistic view of the effects of small molecules in biological systems. Equally important is the ability for non-experts to access and utilize state of the art cheminformatics method and models. In this review we present recent developments in cheminformatics methodologies and infrastructure that provide a robust, distributed approach to mining large and complex chemical datasets. In the area of methodology development, we highlight recent work on characterizing structure-activity landscapes, Quantitative Structure Activity Relationship (QSAR) model domain applicability and the use of chemical similarity in text mining. In the area of infrastructure, we discuss a distributed web services framework that allows easy deployment and uniform access to computational (statistics, cheminformatics and computational chemistry) methods, data and models. We also discuss the development of PubChem derived databases and highlight techniques that allow us to scale the infrastructure to extremely large compound collections, by use of distributed processing on Grids. Given that the above work is applicable to arbitrary types of cheminformatics problems, we also present some case studies related to virtual screening for anti-malarials and predictions of anti-cancer activity.

The interplay between evolution, regulation and tissue specificity in the Human Hereditary Diseasome

BackgroundHuman disease genes can be distinguished from essential (embryonically lethal) and non-disease genes using gene attributes. Such attributes include gene age, tissue specificity of expression, regulatory capacity, sequence length, rate of sequence variation and capacity for interaction. The resulting information has been used to inform data mining approaches seeking to identify novel disease genes. Given the dynamic nature of this field and the rapid rise in relevant information, we have chosen to perform a single integrated mining approach to explore relationships among gene attributes and thereby characterise evolutionary trends associated with disease genes.ResultsAll against all cross comparison of 2,522 disease gene attributes revealed significant relationships existed between the age, disease-association and expression pattern of genes and the tissues within which they are expressed. We found that the over-representation of disease genes among old genes holds for tissue-specific genes, but the correlation between age and disease association vanished when conditioning on tissue-specificity. Of the 32 tissues studied, the genes expressed in pancreas are on average older than the genes expressed in any other tissue, while the testis expressed the lowest proportion of old genes. Following a focussed analysis on the impact of regulatory apparatus on evolution of disease genes, we show that regulators, comprising transcription factors and post-translation modified proteins, are over-represented among ancient disease genes. In addition, we show that the proportion of regulator genes is affected by gene age among disease genes and by tissue-specificity among non-disease genes. Finally, using 55,606 true positive gene interaction data, we find that old disease genes interacts with other old disease genes and interacting new genes interacts with genes originating from higher phylostrata.ConclusionThis study supports the non-random nature of the human diseasome. We have identified a variety of distinct features and correlations to other molecular attributes that can be used to distinguish the set of disease causing genes. This was achieved by harnessing the power of mining large scale datasets from OMIM and other databases. Ultimately such knowledge may contribute to the identification of novel human disease genes and an enhanced understanding of human biology.

Mining large heterogeneous data sets in drug discovery

Background: Increasingly, effective drug discovery involves the searching and data mining of large volumes of information from many sources covering the domains of chemistry, biology and pharmacology amongst others. This has led to a proliferation of databases and data sources relevant to drug discovery. Objective: This paper provides a review of the publicly-available large-scale databases relevant to drug discovery, describes the kinds of data mining approaches that can be applied to them and discusses recent work in integrative data mining that looks for associations that pan multiple sources, including the use of Semantic Web techniques. Conclusion: The future of mining large data sets for drug discovery requires intelligent, semantic aggregation of information from all of the data sources described in this review, along with the application of advanced methods such as intelligent agents and inference engines in client applications.

A Selection Bias Conflict and Frequentist versus Bayesian Viewpoints

In many branches of modern science, researchers first study or mine large datasets, and then select the parameters they estimate and the data they use and publish. Such data-based selection complicates formal statistical inference. An example discussed here for the purpose of illustration, is that of pharmaceutical companies that typically conduct many experiments but may publish only selected data. The selection often depends on the outcomes of the experiments since naturally there is interest in potentially useful drugs, and it is in general unclear how it should affect inference. Is this effect the same for the company and the public? Does it matter if they are Bayesian or frequentist? Should the company reveal all experiments it conducts and, if so, how should this change the conclusions?This note discusses these questions in terms of a simple example of a sequence of binomial experiments conducted by a pharmaceutical company, where results are published only if the number of “failures” is small. We do not suggest that this example corresponds to reality in the pharmaceutical industry, nor in science in general; our goal is to elaborate on the importance and difficulties of taking selection into account when performing statistical analysis.

Bellwether analysis

How to mine massive datasets is a challenging problem with great potential value. Motivated by this challenge, much effort has concentrated on developing scalable versions of machine learning algorithms. However, the cost of mining large datasets is not just computational; preparing the datasets into the “right form” so that learning algorithms can be applied is usually costly, due to the human labor that is typically required and a large number of choices in data preparation, which include selecting different subsets of data and aggregating data at different granularities. We make the key observation that, for a number of practically motivated problems, these choices can be defined using database queries and analyzed in an automatic and systematic manner. Specifically, we propose a new class of data-mining problem, called bellwether analysis , in which the goal is to find a few query-defined predictors (e.g., first week sales of Peoria, IL of an item) that can be used to accurately predict the result of a target query (e.g., first year worldwide sales of the item) from a large number of queries that define candidate predictors. To make a prediction for a new item, the data needed to generate such predictors has to be collected (e.g., selling the new item in Peoria, IL for a week and collecting the sales data). A useful predictor is one that has high prediction accuracy and a low data-collection cost. We call such a cost-effective predictor a bellwether . This article introduces bellwether analysis, which integrates database query processing and predictive modeling into a single framework, and provides scalable algorithms for large datasets that cannot fit in main memory. Through a series of extensive experiments, we show that bellwethers do exist in real-world databases, and that our computation techniques achieve good efficiency on large datasets.

Discovery of feature-based hot spots using supervised clustering

Feature-based hot spots are localized regions where the attributes of objects attain high values. There is considerable interest in automatic identification of feature-based hot spots. This paper approaches the problem of finding feature-based hot spots from a data mining perspective, and describes a method that relies on supervised clustering to produce a list of hot spot regions. Supervised clustering uses a fitness function rewarding isolation of the hot spots to optimally subdivide the dataset. The clusters in the optimal division are ranked using the interestingness of clusters that encapsulate their utility for being hot spots. Hot spots are associated with the top ranked clusters. The effectiveness of supervised clustering as a hot spot identification method is evaluated for four conceptually different clustering algorithms using a dataset describing the spatial distribution of ground ice on Mars. Clustering solutions are visualized by specially developed raster approximations. Further assessment of the ability of different algorithms to yield hot spots is performed using raster approximations. Density-based clustering algorithm is found to be the most effective for hot spot identification. The results of the hot spot discovery by supervised clustering are comparable to those obtained using the G * statistic, but the new method offers a high degree of automation, making it an ideal tool for mining large datasets for the existence of potential hot spots.

Algorithms for Storytelling

We formulate a new data mining problem called storytelling as a generalization of redescription mining. In traditional redescription mining, we are given a set of objects and a collection of subsets defined over these objects. The goal is to view the set system as a vocabulary and identify two expressions in this vocabulary that induce the same set of objects. Storytelling, on the other hand, aims to explicitly relate object sets that are disjoint (and hence, maximally dissimilar) by finding a chain of (approximate) redescriptions between the sets. This problem finds applications in bioinformatics, for instance, where the biologist is trying to relate a set of genes expressed in one experiment to another set, implicated in a different pathway. We outline an efficient storytelling implementation that embeds the CARTwheels redescription mining algorithm in an A* search procedure, using the former to supply next move operators on search branches to the latter. This approach is practical and effective for mining large datasets and, at the same time, exploits the structure of partitions imposed by the given vocabulary. Three application case studies are presented: a study of word overlaps in large English dictionaries, exploring connections between genesets in a bioinformatics dataset, and relating publications in the PubMed index of abstracts.

Construction, Visualisation, and Clustering of Transcription Networks from Microarray Expression Data

Network analysis transcends conventional pairwise approaches to data analysis as the context of components in a network graph can be taken into account. Such approaches are increasingly being applied to genomics data, where functional linkages are used to connect genes or proteins. However, while microarray gene expression datasets are now abundant and of high quality, few approaches have been developed for analysis of such data in a network context. We present a novel approach for 3-D visualisation and analysis of transcriptional networks generated from microarray data. These networks consist of nodes representing transcripts connected by virtue of their expression profile similarity across multiple conditions. Analysing genome-wide gene transcription across 61 mouse tissues, we describe the unusual topography of the large and highly structured networks produced, and demonstrate how they can be used to visualise, cluster, and mine large datasets. This approach is fast, intuitive, and versatile, and allows the identification of biological relationships that may be missed by conventional analysis techniques. This work has been implemented in a freely available open-source application named BioLayout Express 3D.

An interactive approach to mining gene expression data

Effective identification of coexpressed genes and coherent patterns in gene expression data is an important task in bioinformatics research and biomedical applications. Several clustering methods have recently been proposed to identify coexpressed genes that share similar coherent patterns. However, there is no objective standard for groups of coexpressed genes. The interpretation of co-expression heavily depends on domain knowledge. Furthermore, groups of coexpressed genes in gene expression data are often highly connected through a large number of "intermediate" genes. There may be no clear boundaries to separate clusters. Clustering gene expression data also faces the challenges of satisfying biological domain requirements and addressing the high connectivity of the data sets. In this paper, we propose an interactive framework for exploring coherent patterns in gene expression data. A novel coherent pattern index is proposed to give users highly confident indications of the existence of coherent patterns. To derive a coherent pattern index and facilitate clustering, we devise an attraction tree structure that summarizes the coherence information among genes in the data set. We present efficient and scalable algorithms for constructing attraction trees and coherent pattern indices from gene expression data sets. Our experimental results show that our approach is effective in mining gene expression data and is scalable for mining large data sets.

A Data-Driven Multidimensional Indexing Method for Data Mining in Astrophysical Databases

Large archives and digital sky surveys with dimensions of bytes currently exist, while in the near future they will reach sizes of the order of . Numerical simulations are also producing comparable volumes of information. Data mining tools are needed for information extraction from such large datasets. In this work, we propose a multidimensional indexing method, based on a static R-tree data structure, to efficiently query and mine large astrophysical datasets. We follow a top-down construction method, called VAMSplit, which recursively splits the dataset on a near median element along the dimension with maximum variance. The obtained index partitions the dataset into nonoverlapping bounding boxes, with volumes proportional to the local data density. Finally, we show an application of this method for the detection of point sources from a gamma-ray photon list.

Improving ecological niche models by data mining large environmental datasets for surrogate models

WhyWhere is a new ecological niche modeling (ENM) algorithm for mapping and explaining the distribution of species. The algorithm uses image processing methods to efficiently sift through large amounts of data to find the few variables that best predict species occurrence. The purpose of this paper is to describe and justify the main parameterizations and to show preliminary success at rapidly providing accurate, scalable, and simple ENMs. Preliminary results for six species of plants and animals in different regions indicate a significant ( p < 0.01) 14% increase in accuracy over the GARP algorithm using models with few, typically two, variables. The increase is attributed to access to additional data, particularly remotely sensed monthly versus annual climate averages. WhyWhere is also six times faster than GARP on large datasets. A data mining based approach with transparent access to remote data archives is a new paradigm for ENM, particularly suited to finding correlates in large databases of fine resolution surfaces. Software for WhyWhere is freely available, both as a service and in a desktop downloadable form from the web site http://biodi.sdsc.edu/ww_home.html.

A role for Pareto optimality in mining performance data

Improvements in performance modeling and identification of computational regimes within software libraries is a critical first step in developing software libraries that are truly agile with respect to the application as well as to the hardware. It is shown here that Pareto ranking, a concept from multi-objective optimization, can be an effective tool for mining large performance datasets. The approach is illustrated using software performance data gathered using both the public domain LAPACK library and an asynchronous communication library based on IBM LAPI active message library.

P-autoclass: scalable parallel clustering for mining large data sets

Data clustering is an important task in the area of data mining. Clustering is the unsupervised classification of data items into homogeneous groups called clusters. Clustering methods partition a set of data items into clusters, such that items in the same cluster are more similar to each other than items in different clusters according to some defined criteria. Clustering algorithms are computationally intensive, particularly when they are used to analyze large amounts of data. A possible approach to reduce the processing time is based on the implementation of clustering algorithms on scalable parallel computers. This paper describes the design and implementation of P-AutoClass, a parallel version of the AutoClass system based upon the Bayesian model for determining optimal classes in large data sets. The P-AutoClass implementation divides the clustering task among the processors of a multicomputer so that each processor works on its own partition and exchanges intermediate results with the other processors. The system architecture, its implementation, and experimental performance results on different processor numbers and data sets are presented and compared with theoretical performance. In particular, experimental and predicted scalability and efficiency of P-AutoClass versus the sequential AutoClass system are evaluated and compared.

The Icelandic Healthcare Database and informed consent.

Modern information technology is rapidly changing how new knowledge is discovered in most fields of science, including medicine. This technology offers interesting possibilities in the development of methods to understand diseases better, but it also presents new ethical challenges. The new technology offers the possibility of mining large data sets for knowledge, without a priori hypotheses, by systematically juxtaposing various data in the search for the best fit. This kind of pure combinatorial analysis may be particularly powerful in the case of the common diseases, most of which are complex and have remained beyond the reach of the classic hypothesis-driven . . .

Intelligent simulation tools for mining large scientific data sets

This paper describes problems, challenges, and opportunities forintelligent simulation of physical systems. Prototype intelligent simulation tools have been constructed for interpreting massive data sets from physical fields and for designing engineering systems. We identify the characteristics of intelligent simulation and describe several concrete application examples. These applications, which include weather data interpretation, distributed control optimization, and spatio-temporal diffusion-reaction pattern analysis, demonstrate that intelligent simulation tools are indispensable for the rapid prototyping of application programs in many challenging scientific and engineering domains.

Set-oriented data mining in relational databases

Data mining is an important real-life application for businesses. It is critical to find efficient ways of mining large data sets. In order to benefit from the experience with relational databases, a set-oriented approach to mining data is needed. In such an approach, the data mining operations are expressed in terms of relational or set-oriented operations. Query optimization technology can then be used for efficient processing. In this paper, we describe set-oriented algorithms for mining association rules. Such algorithms imply performing multiple joins and thus may appear to be inherently less efficient than special-purpose algorithms. We develop new algorithms that can be expressed as SQL queries, and discuss optimization of these algorithms. After analytical evaluation, an algorithm named SETM emerges as the algorithm of choice. Algorithm SETM uses only simple database primitives, viz., sorting and merge-scan join. Algorithm SETM is simple, fast, and stable over the range of parameter values. It is easily parallelized and we suggest several additional optimizations. The set-oriented nature of Algorithm SETM makes it possible to develop extensions easily and its performance makes it feasible to build interactive data mining tools for large databases.