Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Data mining”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3Linked to original sources
PubMed2000

Visual management of large scale data mining projects.

This paper describes a unified framework for visualizing the preparations for, and results of, hundreds of machine learning experiments. These experiments were designed to improve the accuracy of enzyme functional predictions from sequence, and in many cases were successful. Our system provides graphical user interfaces for defining and exploring training datasets and various representational alternatives, for inspecting the hypotheses induced by various types of learning algorithms, for visualizing the global results, and for inspecting in detail results for specific training sets (functions) and examples (proteins). The visualization tools serve as a navigational aid through a large amount of sequence data and induced knowledge. They provided significant help in understanding both the significance and the underlying biological explanations of our successes and failures. Using these visualizations it was possible to efficiently identify weaknesses of the modular sequence representations and induction algorithms which suggest better learning strategies. The context in which our data mining visualization toolkit was developed was the problem of accurately predicting enzyme function from protein sequence data. Previous work demonstrated that approximately 6% of enzyme protein sequences are likely to be assigned incorrect functions on the basis of sequence similarity alone. In order to test the hypothesis that more detailed sequence analysis using machine learning techniques and modular domain representations could address many of these failures, we designed a series of more than 250 experiments using information-theoretic decision tree induction and naive Bayesian learning on local sequence domain representations of problematic enzyme function classes. In more than half of these cases, our methods were able to perfectly discriminate among various possible functions of similar sequences. We developed and tested our visualization techniques on this application.

Alcohol Dehydrogenase↗
PubMed2025

Integrating explainable AI with multiomics systems biology and EHR data mining for personalized drug repurposing in Alzheimer's disease.

Alzheimer's disease (AD) is characterized by region- and patient-specific molecular heterogeneity, which hinders therapeutic design. In this study, we introduce PRISM-ML (PRecision-medicine using Interpretable Systems and Multiomics with Machine Learning), an open-source integrated analysis pipeline that combines interpretable machine learning with systems biology and electronic health record (EHR) data mining to elucidate the molecular diversity of AD and predict promising drug repurposing opportunities. First, we integrated and harmonized transcriptomic (bulk RNA-seq) and genomic (genome-wide association study) data from 2105 brain samples, each with matched data from the same individual (1363 AD patients, 742 controls; nine tissues), sourced from three independent studies. Random forest classifiers with SHapley Additive exPlanations (SHAP) identified patient-specific biomarkers; unsupervised clustering resolved 36 molecularly distinct "subtissues" (clusters of samples); and gene-gene co-expression networks prioritized 262 high-centrality bottleneck genes as putative regulators of dysregulated pathways. Next, knowledge graph-based drug repurposing predicted six FDA-approved drugs that simultaneously target multiple bottleneck genes and multiple AD-relevant pathways. Notably, in a large U.S. de-identified insurance-claims database (n = 364733), exposure to promethazine, one of the candidate drugs, was associated with a 57-62 % lower incidence of AD versus an active antihistamine comparator (adjusted hazard ratio 0.38; inverse-probability weighted 0.43; both p < 0.001), providing real-world support for its repurposing potential. In summary, PRISM-ML, as an explainable multi-omics analysis pipeline, is readily transferable to other complex diseases, advancing precision medicine.

Computational Biology↗
PubMed1998

Knowledge discovery and data mining to assist natural language understanding.

As natural language processing systems become more frequent in clinical use, methods for interpreting the output of these programs become increasingly important. These methods require the effort of a domain expert, who must build specific queries and rules for interpreting the processor output. Knowledge discovery and data mining tools can be used instead of a domain expert to automatically generate these queries and rules. C5.0, a decision tree generator, was used to create a rule base for a natural language understanding system. A general-purpose natural language processor using this rule base was tested on a set of 200 chest radiograph reports. When a small set of reports, classified by physicians, was used as the training set, the generated rule base performed as well as lay persons, but worse than physicians. When a larger set of reports, using ICD9 coding to classify the set, was used for training the system, the rule base performed worse than the physicians and lay persons. It appears that a larger, more accurate training set is needed to increase performance of the method.

Artificial Intelligence↗
PubMedDate not supplied

A data mining approach for analyzing density maps representing macromolecular structures.

Results of electron microscopy-based three-dimensional reconstructions of macromolecules or their complexes are usually stored as density maps. Each point ("voxel") in the map represents a density value and one approach for studying details of the map is to display an isosurface enclosing areas of interest. We have taken a data mining approach not only focusing on the areas of immediate interest but determining all possible separate entities ("blobs") from a density map. After the entire density map is analyzed with our mining program BLOBBER, properties of all detected blobs can be browsed and sets of blobs can be visualized using our VIZBLOB program. Since BLOBBER analyzes density maps using only density information and relates it to spatial relationships, BLOBBER can be used to analyze symmetrical or asymmetrical density maps from any source. To test our program we have analyzed published bacteriophage PRD1 reconstructions. We identified various structural details ranging from individual proteins to major complexes such as the whole capsid shell and more elaborate details of possible connections between membrane interfaces. This approach can also be a useful preprocessing tool for visualizing reconstructions.

Bacteriophages↗
PubMed2000

Fullerene data mining using bibliometrics and database tomography

Database tomography (DT) is a textual database analysis system consisting of two major components: (1) algorithms for extracting multiword phrase frequencies and phrase proximities (physical closeness of the multiword technical phrases) from any type of large textual database, to augment (2) interpretative capabilities of the expert human analyst. DT was used to derive technical intelligence from a fullerenes database derived from the Science Citation Index and the Engineering Compendex. Phrase frequency analysis by the technical domain experts provided the pervasive technical themes of the fullerenes database, and phrase proximity analysis provided the relationships among the pervasive technical themes. Bibliometric analysis of the fullerenes literature supplemented the DT results with author/journal/institution publication and citation data. Comparisons of fullerenes results with past analyses of similarly structured near-earth space, chemistry, hypersonic/supersonic flow, aircraft, and ship hydrodynamics databases are made. One important finding is that many of the normalized bibliometric distribution functions are extremely consistent across these diverse technical domains and could reasonably be expected to apply to broader chemical topics than fullerenes that span multiple structural classes. Finally, lessons learned about integrating the technical domain experts with the data mining tools are presented.

Journal Article↗
PubMed1997

Data mining for regulatory elements in yeast genome.

We have examined methods and developed a general software tool for finding and analyzing combinations of transcription factor binding sites that occur relatively often in gene upstream regions (putative promoter regions) in the yeast genome. Such frequently occurring combinations may be essential parts of possible promoter classes. The regions upstream to all genes were first isolated from the yeast genome database MIPS using the information in the annotation files of the database. The ones that do not overlap with coding regions were chosen for further studies. Next, all occurrences of the yeast transcription factor binding sites, as given in the IMD database, were located in the genome and in the selected regions in particular. Finally, by using a general purpose data mining software in combination with our own software, which parametrizes the search, we can find the combinations of binding sites that occur in the upstream regions more frequently than would be expected on the basis of the frequency of individual sites. The procedure also finds so-called association rules present in such combinations. The developed tool is available for use through the WWW.

Binding Sites↗
PubMed1998

The Merck Gene Index browser: an extensible data integration system for gene finding, gene characterization and EST data mining.

MOTIVATION: To make effective use of the vast amounts of expressed sequence tag (EST) sequence data generated by the Merck-sponsored EST project and other similar efforts, sequences must be organized into gene classes, and scientists must be able to 'mine' the gene class data in the context of related genomic data. RESULTS: This paper presents the Merck Gene Index browser, an easily extensible, World Wide Web-based system for mining the Merck Gene Index (MGI) and related genomic data. The MGI is a non-redundant set of clones and sequences, each representing a distinct gene, constructed from all high-quality 3' EST sequences generated by the Merck-sponsored EST project. The MGI browser integrates data from a variety of sources and storage formats, both local and remote, using an eclectic integration strategy, including a federation of relational databases, a local data warehouse and simple hypertext links. Data currently integrated include: LENS cDNA clone and EST data, dbEST protein and non-EST nucleic acid similarity data, WashU sequence chromatograms. Entrez sequence and Medline entries, and UniGene gene clusters. Flatfile sequence data are accessed using the Bioapps server, an internally developed client-server system that supports generic sequence analysis applications. Browser data are retrieved and formatted by means of the Bioinformatics Data Integration Toolkit (B-DIT), a new suite of Perl routines.

Abstracting and Indexing↗
PubMed1999

Prospecting for gold in the data mine.

As a transaction based industry, health care is data rich. A new generation of business-supporting information technology is emerging that can transform such data into knowledge critical to sustain value in health care. In order to adopt successfully this new information technology, physicians and other health care leaders must come to understand the value of complete, accurate and consistent coding of clinical activities. The clinical laboratory has a pervasive role in health care. With its recent federally assigned responsibility to assure clinically relevant testing through ICD-9-CM and CPT coding, and its experience with computerized information systems, the clinical laboratory is in an ideal position to become a champion of the new information technology.

Clinical Laboratory Information Systems↗
PubMed1997

Data mining of inputs: analysing magnitude and functional measures.

The problem of data encoding and feature selection for training back-propagation neural networks is well known. The basic principles are to avoid encrypting the underlying structure of the data, and to avoid using irrelevant inputs. This is not easy in the real world, where we often receive data which has been processed by at least one previous user. The data may contain too many instances of some class, and too few instances of other classes. Real data sets often include many irrelevant or redundant input fields. This paper examines the use of weight matrix analysis techniques and functional measures using two real (and hence noisy) data sets. The first part of this paper examines the use of the weight matrix of the trained neural network itself to determine which inputs are significant. A new technique is introduced and compared with two other techniques from the literature. We present our experience and results on some satellite data augmented by a terrain model. The task was to predict the forest supra-type based on the available information. A brute force technique eliminating randomly selected inputs was used to validate our approach. The second part of this paper examines the use of measures to determine the functional contribution of inputs to outputs. Inputs which include minor but unique information to the network are more significant than inputs with higher magnitude contribution but providing redundant information, which is also provided by another input. A comparison is made to sensitivity analysis, where the sensitivity of outputs to input perturbation is used as a measure of the significance of inputs. This paper presents a novel functional analysis of the weight matrix based on a technique developed for determining the behavioral significance of hidden neurons. This is compared with the application of the same technique to the training and test data. Finally, a novel aggregation technique is introduced.

Algorithms↗
PubMed2000

Visualization of multiple influences on ocellar flight control in giant honeybees with the data-mining tool Viscovery SOMine.

Viscovery SOMine is a software tool for advanced analysis and monitoring of numerical data sets. It was developed for professional use in business, industry, and science and to support dependency analysis, deviation detection, unsupervised clustering, nonlinear regression, data association, pattern recognition, and animated monitoring. Based on the concept of self-organizing maps (SOMs), it employs a robust variant of unsupervised neural networks--namely, Kohonen's Batch-SOM, which is further enhanced with a new scaling technique for speeding up the learning process. This tool provides a powerful means by which to analyze complex data sets without prior statistical knowledge. The data representation contained in the trained SOM is systematically converted to be used in a spectrum of visualization techniques, such as evaluating dependencies between components, investigating geometric properties of the data distribution, searching for clusters, or monitoring new data. We have used this software tool to analyze and visualize multiple influences of the ocellar system on free-flight behavior in giant honeybees. Occlusion of ocelli will affect orienting reactivities in relation to flight target, level of disturbance, and position of the bee in the flight chamber; it will induce phototaxis and make orienting imprecise and dependent on motivational settings. Ocelli permit the adjustment of orienting strategies to environmental demands by enforcing abilities such as centering or flight kinetics and by providing independent control of posture and flight course.

Animals↗
PubMed2000

Discovery of predictive models in an injury surveillance database: an application of data mining in clinical research.

A new, evolutionary computation-based approach to discovering prediction models in surveillance data was developed and evaluated. This approach was operationalized in EpiCS, a type of learning classifier system specially adapted to model clinical data. In applying EpiCS to a large, prospective injury surveillance database, EpiCS was found to create accurate predictive models quickly that were highly robust, being able to classify > 99% of cases early during training. After training, EpiCS classified novel data more accurately (p < 0.001) than either logistic regression or decision tree induction (C4.5), two traditional methods for discovering or building predictive models.

Artificial Intelligence↗
PubMed1998

Data mining by clinicians.

Clinical databases are becoming commonplace in healthcare environments. However, clinicians have been unable to readily explore these information sources, because currently available data retrieval tools require substantial technical skill, as well as knowledge of the underlying database structures. To address this, we have defined a group of "atomic" queries, including both population-based and temporal predicates, to enable the extraction of clinically meaningful information form these databases. DXtractor is an application that incorporates this functionality, and allows clinicians to simply combine these atomic queries. In doing so, arbitrarily complex data retrieval and exploration becomes possible for the non-programming clinician.

Databases as Topic↗
PubMed1998

GeneCards: a novel functional genomics compendium with automated data mining and query reformulation support.

MOTIVATION: Modern biology is shifting from the 'one gene one postdoc' approach to genomic analyses that include the simultaneous monitoring of thousands of genes. The importance of efficient access to concise and integrated biomedical information to support data analysis and decision making is therefore increasing rapidly, in both academic and industrial research. However, knowledge discovery in the widely scattered resources relevant for biomedical research is often a cumbersome and non-trivial task, one that requires a significant amount of training and effort. RESULTS: To develop a model for a new type of topic-specific overview resource that provides efficient access to distributed information, we designed a database called 'GeneCards'. It is a freely accessible Web resource that offers one hypertext 'card' for each of the more than 7000 human genes that currently have an approved gene symbol published by the HUGO/GDB nomenclature committee. The presented information aims at giving immediate insight into current knowledge about the respective gene, including a focus on its functions in health and disease. It is compiled by Perl scripts that automatically extract relevant information from several databases, including SWISS-PROT, OMIM, Genatlas and GDB. Analyses of the interactions of users with the Web interface of GeneCards triggered development of easy-to-scan displays optimized for human browsing. Also, we developed algorithms that offer 'ready-to-click' query reformulation support, to facilitate information retrieval and exploration. Many of the long-term users turn to GeneCards to quickly access information about the function of very large sets of genes, for example in the realm of large-scale expression studies using 'DNA chip' technology or two-dimensional protein electrophoresis. AVAILABILITY: Freely available at http://bioinformatics.weizmann.ac.il/cards/ CONTACT: cards@bioinformatics.weizmann.ac.il

Algorithms↗