Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

BIOE 402. Medical Technology Assessment. 2 or 3 hours. Bioentrepreneur course. Assessment of medical technology in the context of commercialization. Objectives, competition, market share, funding, pricing, manufacturing, growth, and intellectual property; many issues unique to biomedical products. Course Information: 2 undergraduate hours. 3 graduate hours. Prerequisite(s): Junior standing or above and consent of the instructor.

“Bioinformatics” 특집을 내면서

Access to unified datasets of protein and genetic interactions is critical for interrogation of gene/protein function and analysis of global network properties. BioGRID is a freely accessible database of physical and genetic interactions available at http://www.thebiogrid.org. BioGRID release version 2.0 includes >116 000 interactions from Saccharomyces cerevisiae, Caenorhabditis elegans, Drosophila melanogaster and Homo sapiens. Over 30 000 interactions have recently been added from 5778 sources through exhaustive curation of the Saccharomyces cerevisiae primary literature. An internally hyper-linked web interface allows for rapid search and retrieval of interaction data. Full or user-defined datasets are freely downloadable as tab-delimited text files and PSI-MI XML. Pre-computed graphical layouts of interactions are available in a variety of file formats. User-customized graphs with embedded protein, gene and interaction attributes can be constructed with a visualization system called Osprey that is dynamically linked to the BioGRID.

/pdf/biogrid-a-general-repository-for-interaction-datasets-3ul1ct2q6g.pdf

BioGRID: a general repository for interaction datasets

Functional partnerships between proteins are at the core of complex cellular phenotypes, and the networks formed by interacting proteins provide researchers with crucial scaffolds for modeling, data reduction and annotation. STRING is a database and web resource dedicated to protein–protein interactions, including both physical and functional interactions. It weights and integrates information from numerous sources, including experimental repositories, computational prediction methods and public text collections, thus acting as a metadatabase that maps all interaction evidence onto a common set of genomes and proteins. The most important new developments in STRING 8 over previous releases include a URL-based programming interface, which can be used to query STRING from other resources, improved interaction prediction via genomic neighborhood in prokaryotes, and the inclusion of protein structures. Version 8.0 of STRING covers about 2.5 million proteins from 630 organisms, providing the most comprehensive view on protein–protein interactions currently available. STRING can be reached at http://string-db.org/.

/pdf/string-8-a-global-view-on-proteins-and-their-functional-56hd0uvpxl.pdf

STRING 8—a global view on proteins and their functional interactions in 630 organisms

Mapping the pathways that give rise to metastasis is one of the key challenges of breast cancer research. Recently, several large-scale studies have shed light on this problem through analysis of gene expression profiles to identify markers correlated with metastasis. Here, we apply a protein-network-based approach that identifies markers not as individual genes but as subnetworks extracted from protein interaction databases. The resulting subnetworks provide novel hypotheses for pathways involved in tumor progression. Although genes with known breast cancer mutations are typically not detected through analysis of differential expression, they play a central role in the protein network by interconnecting many differentially expressed genes. We find that the subnetwork markers are more reproducible than individual marker genes selected without network information, and that they achieve higher accuracy in the classification of metastatic versus non-metastatic tumors.

/pdf/network-based-classification-of-breast-cancer-metastasis-41ka7eod3b.pdf

Network-based classification of breast cancer metastasis.

The Biomolecular Interaction Network Database (BIND) (http://bind.ca) archives biomolecular interaction, reaction, complex and pathway information. Our aim is to curate the details about molecular interactions that arise from published experimental research and to provide this information, as well as tools to enable data analysis, freely to researchers worldwide. BIND data are curated into a comprehensive machine-readable archive of computable information and provides users with methods to discover interactions and molecular mechanisms. BIND has worked to develop new methods for visualization that amplify the underlying annotation of genes and proteins to facilitate the study of molecular interaction networks. BIND has maintained an open database policy since its inception in 1999. Data growth has proceeded at a tremendous rate, approaching over 100 000 records. New services provided include a new BIND Query and Submission interface, a Standard Object Access Protocol service and the Small Molecule Interaction Database (http://smid.blueprint.org) that allows users to determine probable small molecule binding sites of new sequences and examine conserved binding residues.

The Biomolecular Interaction Network Database and related tools 2005 update

The majority of experimentally verified molecular interaction and biological pathway data are present in the unstructured text of biomedical journal articles where they are inaccessible to computational methods. The Biomolecular interaction network database (BIND) seeks to capture these data in a machine-readable format. We hypothesized that the formidable task-size of backfilling the database could be reduced by using Support Vector Machine technology to first locate interaction information in the literature. We present an information extraction system that was designed to locate protein-protein interaction data in the literature and present these data to curators and the public for review and entry into BIND. Cross-validation estimated the support vector machine's test-set precision, accuracy and recall for classifying abstracts describing interaction information was 92%, 90% and 92% respectively. We estimated that the system would be able to recall up to 60% of all non-high throughput interactions present in another yeast-protein interaction database. Finally, this system was applied to a real-world curation problem and its use was found to reduce the task duration by 70% thus saving 176 days. Machine learning methods are useful as tools to direct interaction and pathway database back-filling; however, this potential can only be realized if these techniques are coupled with human review and entry into a factual database such as BIND. The PreBIND system described here is available to the public at http://bind.ca
 . Current capabilities allow searching for human, mouse and yeast protein-interaction information.

/pdf/prebind-and-textomy-mining-the-biomedical-literature-for-3x8qel20sa.pdf

PreBIND and Textomy - mining the biomedical literature for protein-protein interactions using a support vector machine

We present a method to generate glyphs which convey complex information in graphical form. A glyph has a linear geometry which is specified using geometric operations, each represented by characters nested in a string. This format allows several glyph strings to be concatenated, resulting in more complex geometries. We explore automatic generation of a large number of glyphs using a genetic algorithm. To measure the visual distinctness between two glyph geometries, we use the iterative closest point algorithm. We apply these methods to create two different types of representations for biological proteins, transforming the rich data describing their various characteristics into graphical form. The representations are automatically built from a finite set of glyphs, which have been created manually or using the genetic algorithm.

/pdf/generation-of-glyphs-for-conveying-complex-information-with-7bghhf2bxe.pdf

Brigitte Tuekam

Papers

The Biomolecular Interaction Network Database and related tools 2005 update

PreBIND and Textomy - mining the biomedical literature for protein-protein interactions using a support vector machine

Generation of glyphs for conveying complex information, with application to protein representations

Generation of glyphs for conveying complex information, with application to protein representations