GeneMANIA prediction server 2013 update.

Zuberi K, Franz M, Rodriguez H, Montojo J, Lopes CT, Bader GD, Morris Q. · 2013

Read GeneMANIA prediction server 2013 update. by Zuberi K, Franz M, Rodriguez H, Montojo J, Lopes CT, Bader GD, Morris Q. (2013) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc3692113.txt.

Introduction

Zuberi and colleagues update the GeneMANIA prediction server, a web tool that guesses gene function from network data. Given a list of query genes, the server extends it with genes that behave similarly across large genomics and proteomics datasets, and it reports how much each dataset contributed to the prediction.

GeneMANIA also works in reverse. Given one gene of unknown function, it finds the small set of genes most likely to share that function. Enriched Gene Ontology terms among them can hint at the answer. Seven organisms are supported, users can choose which datasets to search, and researchers can upload their own data to fold into the analysis.

Catalog record

CREATORZuberi K, Franz M, Rodriguez H, Montojo J, Lopes CT, Bader GD, Morris Q.
YEAR2013
COLLECTIONPAPERS
FORMATTXT
FILE SIZE35 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC3692113

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

GeneMANIA Prediction Server 2013 Update By Khalid Zuberi, Max Franz, Harold Rodriguez, Jason Montojo, Christian Tannus Lopes, Gary D Bader, Quaid Morris Published in: Nucleic Acids Research DOI: 10.1093/nar/gkt533 ABSTRACT Abstract GeneMANIA (http://www.genemania.org) is a flexible user-friendly web interface for generating hypotheses about gene function, analyzing gene lists and prioritizing genes for functional assays. Given a query gene list, GeneMANIA extends the list with functionally similar genes that it identifies using available genomics and proteomics data. GeneMANIA also reports weights that indicate the predictive value of each selected data set for the query. GeneMANIA can also be used in a function prediction setting: given a query gene, GeneMANIA finds a small set of genes that are most likely to share function with that gene based on their interactions with it. Enriched Gene Ontology categories among this set can sometimes point to the function of the gene. Seven organisms are currently supported (Arabidopsis thaliana, Caenorhabditis elegans, Drosophila melanogaster, Mus musculus, Homo sapiens, Rattus norvegicus and Saccharomyces cerevisiae), and hundreds of data sets have been collected from GEO, BioGRID, IRefIndex and I2D, as well as organism-specific functional genomics data sets. Users can customize their search by selecting specific data sets to query and by uploading their own data sets to analyze. INTRODUCTION The GeneMANIA prediction server (http://www.genemania.org) has been in operation for the past 3 years (1) although prototypes with limited functionality were available from 2008 (2). Details about the server and its prediction accuracy on benchmark data sets are available in the original NAR Web server article (1) and associated publications that describe the label propagation algorithm used by GeneMANIA to find related genes (2–4) and the algorithm used to assign weights to networks (2,5) (see Table 1 for details). The prediction server is complemented by the GeneMANIA Cytoscape app (6), which makes similar functionality available for desktop use in addition to a set of unpublished software tools for power users that can be run from the command line that are described at http://pages.genemania.org/tools/. See Table 2 for a list of additional resources associated with GeneMANIA. [Table: GeneMANIA network weighting algorithms] [Table: GeneMANIA associated resources] In this article, we describe updates to the GeneMANIA prediction server and data sets that have occurred over the past 3 years, provide a brief guide to the various resources associated with the GeneMANIA project and answer some of the frequently asked questions about GeneMANIA received from users. In the interest of space, we assume that the reader has already read our previous NAR Web server article (1). GeneMANIA use cases GeneMANIA answers three main types of biological questions by searching diverse network data types, such as physical, predicted and co-expression interactions. Given a single gene, GeneMANIA will find the most closely connected genes. If the function of the query gene is poorly characterized, identifying interacting genes of known function helps predict the function of the unknown gene (7). Given a query list of genes, GeneMANIA will find more genes like those in the list. In this use, GeneMANIA can be thought of as a ‘gene recommender system’ (4), similar to book (or music or movie) recommender systems, which suggest other books (music or movies) that you might like based on your past selections. If the query list is long enough (see the section on network weighting), GeneMANIA can suggest related genes using the types of connections that strongly connect the query genes. For example, if the query genes are part of a protein complex, GeneMANIA will predict additional members of that complex using primarily physical interactions (which highly connect the query genes); or if the query genes are protein kinases, GeneMANIA will suggest other protein kinases by using the protein domain similarity network. Networks are assigned weights based on how relevant they are to the query genes, and non-relevant networks will receive zero weights and not be shown in the results. Finally, GeneMANIA is a general purpose network search engine that supports selecting networks to search out of hundreds of options to identify a set of connections among query genes (the network weighting method should be set to ‘equal by network’ in this case). This is particularly useful when used in the Cytoscape app to find a network of interactions to process further using Cytoscape’s analysis features. GeneMANIA philosophy Don’t get in the way of the data We have attempted to make our interface and algorithm flexible enough that arbitrary gene networks can be uploaded. Although we depend heavily on annotation and curation efforts of others, we try to avoid doing curation or make choices for the user ourselves. This means that by default, we include all network data that we have for the organism with two exceptions: we currently include only the 20 co-expression data sets that are generally most informative for gene function prediction to speed up queries that use the default settings; also, we do not include functional interaction networks generated using Gene Ontology (GO) annotation because they introduce circularity in our calibration and performance measuring algorithms, which currently rely on performing cross-validation with GO annotation. Nonetheless, both of these network types can be added to the query using the advanced options panel. Ease of use We try to choose default behavior so that casual users can get useful information quickly: only two mouse clicks are needed for a gene search (choose an organism, type or paste in the gene and press GO!). This means we have made certain decisions about gene identifier mapping: we only recognize gene names that have a unique correspondence with a gene, and we do not do auto-organism detection because in most cases, gene names are shared among organisms. Much of the complexity of GeneMANIA is hidden in the ‘advanced options panel’ that sits under the gene input box and can be opened to change the default behavior. GENEMANIA UPDATES Since our last publication in 2010, we have made a number of updates to the organisms and data contained within GeneMANIA, to our prediction algorithm and to our user interface. New organisms and data sources We have recently added support for another model organism, rat (Rattus norvegicus), and are developing zebrafish (Danio rerio) and Escherichia coli as our next two model organisms to add. Our data set for the supported organisms is updated approximately twice a year, although we occasionally perform smaller updates when large interaction studies are published. Our front page provides up-to-date statistics on the number of interactions and networks in our database. We perform comparisons between the current version of our database and previous ones to identify data issues and track data set growth (see Table 2). We are unable to make previous data releases available through the Web site, but we do archive them for download, and they can be accessed through our Cytoscape app (6). Our app has the same functionality as the Web site, and queries in the app should give identical results as Web site queries. Web site users can download a JSON file to store their search parameters by selecting the ‘Save search parameters as JSON’ option in the File menu of the results page. This file can be read into our Cytoscape app to exactly reproduce the saved

Read the complete text