Galter Health Sciences Library News - Biosciences

Featured Informatics Resource 11-7-12: FSM Informatics Projects and Research

Wed, 07 Nov 2012 11:44:00 -0500

Today is the final day of Chicago Informatics Week. Actually nine days in all, Informatics Week began on October 30 and ends today, November 7, 2012. All week I’ve been featuring informatics resources in this blog. Today, however, as a final wrap-up, I’m not featuring a resource, but instead focusing a spotlight on some of the many

underway at the Feinberg School of Medicine.

Last week, I attended the Informatics Jam Session hosted by Justin Starren, Chief of the Division of Health and Biomedical Informatics in the Department of Preventive Medicine and Associate Professor in Health and Biomedical Informatics and Medical Social Sciences at FSM. I was impressed by the projects described during these lightning talks, so I decided to feature some of them and other informatics developments growing throughout FSM.

The Department of Medical Social Sciences (MSS) has several noteworthy projects
- The Assessment Center is an online research management site housing a library of assessment tools, storage, and guidance for psychosocial research management.
- Linked to the Assessment center are PROMIS and the NIH Toolbox, both developed by MSS and available as nationwide resources.
- To see a full listing of MSS projects, visit their research page.
Northwestern University Biomedical Informatics Center (NUBIC) has developed numerous tools and solutions to aid informatics research. Some NUBIC resources are:
- REDCap is a secure, web-based application for managing online data capture for research studies.
- The Enterprise Data Warehouse (EDW) houses EHR information from across Northwestern’s health information management systems and also serves as an integrated database of research and clinical data.
- FunDO is a free web tool for managing disease-gene associations based on the NUBIC-developed Disease Ontology. It is a popular resource accessed by users all over the world.
- NUBIC also has a set of developer tools. To see them and view a full listing of NUBIC projects, visit the NUBIC Projects page.
FSM Professor of Pediatrics Marcelo (Bento) Soares is the Director of the Cancer Biology & Epigenomics Program at the Children's Hospital of Chicago Research Center. Dr. Soares is a pioneer in informatics in pediatrics and his lab maintains a set of tools for informatics research.
- Check out Dr. Soares' graph of publications and research at LatticeGrid (a NUBIC-developed tool)!
Keep an eye on the new Institute for Public Health and Medicine. It is developing a graduate program in Health and Biomedical Informatics, and it has numerous centers conducting research in informatics-heavy fields such as epidemiology and biostatistics.

These are just a few of the incredible fields and centers of informatics research at the Feinberg School of Medicine. As Chicago Informatics Week comes to a close, I encourage everyone to keep thinking about informatics beyond Informatics Week and learn more about informatics at FSM and Northwestern University.

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

Featured Informatics Resource 11-6-12: BioMart

Tue, 06 Nov 2012 10:51:00 -0500

Informatics is not always about computation, electronic health records, or molecular manipulation. Sometimes when you are doing research in informatics, all you need is a list of identifiers or accession numbers, or you just need to convert a list of identifiers to another format. Many people don’t know where to turn for such conversion. Today’s featured informatics tool is one of the best for identifier discovery and conversion when searching for Ensembl, UniProt, or Interpro IDs from the European Bioinformatics Institute (EBI):

BioMart

BioMart is part of the suite of tools and databases at the EBI. Other tools are available on the web for identifier conversion, such as DAVID's Gene ID conversion tool, which I mentioned in last Wednesday’s blog. BioMart allows you to start with Entrez/NCBI identifiers, experimental identifiers (eg. Affymetrix) and convert to EBI/Ensembl identifiers, NCBI identifiers and many more.

You can find BioMart under the Databases menu at the EMBL-EBI home page in the Database Browsing category. Let’s say I have a list of mouse Entrez/NCBI gene identifiers, and I want to retrieve all Interpro IDs & descriptions, RefSeq mRNAs, and Ensembl GeneIDs.

Starting with the databases list from the drop-down menu, I choose Ensembl Genes and Mus musculus genes.

Then I choose the Filter. This is the set of identifiers I’m starting with. I’ll choose EntrezGene IDs. I could also choose from many other identifiers, like Affymetrix probe numbers, protein identifiers, etc. Then I paste my list of identifiers in the text box. Note that there are lots of other choices to clearly define your source and list of identifiers.

Now I choose the output from the Attributes menus (by clicking on Attributes in the left menu bar). On this page, Features is automatically highlighted, which means I’ll get a table of my desired output identifiers, but I can use BioMart to retrieve sequences, structures, variation, transcript events, or homologues instead. I’ll choose RefSeq mRNA from the EXTERNAL section, and choose Interpro ID and Interpro Short Description form the PROTEIN DOMAINS section. Now I click the Results button at the top of the left menu bar.

Results may take a while, especially in cases like this when you are retrieving all possible Interpro IDs for a list of genes. You can view your output in the results window, then download the file as TSV, CSV, XLS, HTML, or assembly-specific files.

BioMart gives you a one-stop interface for converting almost any set of identifiers into other database identifiers, retrieving sequences and much more.

Tomorrow’s informatics resource: Not a resource! Tomorrow I will feature many of the fascinating informatics projects underway at the Feinberg School of Medicine.

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

Featured Informatics Resource 11-5-12: Galaxy

Mon, 05 Nov 2012 12:15:00 -0500

Chicago Informatics Week continues. The AMIA Annual Symposium is now underway, and I hope that many Northwestern informatics researchers are attending the conference.

In previous blog posts, I've written about the rise in sequencing experiments in recent years, due to the decrease in costs of sequencing an individual genome. All this sequencing is producing a mountain of data to be analyzed. Many experimenters pay for analysis software or services from the companies that supply the sequencing technology. Other investigators pay a core facility for analysis of sequence data, such as FSM's Bioinformatics Consulting Core. For researchers who want to try analyzing their own data, there are some available options for massive data analysis. Today I feature one such analysis platform:

Galaxy

Galaxy is a web-based open source platform for data-intensive biological research. It is a the product of a collaboration between departments at Penn State and Emory universities, and the tools are hosted at Penn State. The project was funded by numerous grants including awards from the NSF and the NHGRI.

Galaxy has bioinformatics tools for fetching and creating alignments, retrieving sequences, phenotype association and motif discovery; and statistical tools from the R statistical package for multivariate analysis, multiple regression, and more. Galaxy also has a suite of next generation sequencing (NGS) tools for quality control, genome mapping, variant detection, peak calling, indel analysis, and more.

NGS analysis tools are not always easy to use outside of having familiarity with coding for R or without having your own dedicated server for computationally intensive analysis, so Galaxy provides another way to use these tools without needing as much code-writing experience or computational space.

The Galaxy Project home page offers 3 ways of using Galaxy:

Use Galaxy tools on their free public server
Download the source code to install Galaxy locally on your own machine / server
Download source code to run Galaxy on a cloud computing infrastructure (primarily the Amazon Elastic Compute Cloud)

Galaxy also has shared data libraries, workflows and visualizations from users who have completed projects on Galaxy. It also includes an active Wiki site, which is a good place to learn more about Galaxy. There is also a Galaxy Q & A page and a large number of video tutorials available in the Help section of the Galaxy home page.

The use of Galaxy still does require some knowledge of statistics (to select the correct analysis tool) and some knowledge of computers to set up a local or cloud instance of the software. Thanks to comprehensive packages like Galaxy, though, creating a workflow for sequence analysis is now possible for many researchers.

Tomorrow's featured informatics resource: BioMart

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

Featured Informatics Resource 11-2-12: Genome Browsers

Fri, 02 Nov 2012 12:54:00 -0500

Happy Day 4 of Chicago Informatics Week!

Today’s featured informatics resource is actually a set of resources. With the increase in sequencing protocols for clinical and basic science research, these featured resources are ever-more valuable tools for getting information about a genomic region. Today’s feature:

Genome Browsers

There are three major genome browsers used by researchers, but two of them are most heavily used. They are: the Ensembl Genome Browser, the UCSC Genome Browser, and NCBI’s MapViewer.

Ensembl and UCSC Genome Browsers are probably more popular than NCBI’s Map Viewer, but each of these browsers has features and functionality that are unique. Many researchers have their favorite among these tools. We have a basic introductory guide to these three genome browsers available on the Galter website.

What are genome browsers used for?

Genome browsers are valuable for viewing information about your gene of interest in a wider genomic context: see what other genes are nearby
If you run sequencing studies, genome browsers help you to view features in regions of the genome where your sequence tags are mapped
Genome browsers allow you to control the annotation of features when viewing your region of interest: you can view SNPs, annotations in OMIM, get links to translated protein products, and much more
Genome browsers facilitate comparison between or across species to get a picture of synteny

The Ensembl Genome Browser

The Ensembl Genome Browser has many species’ genomes to investigate, more than any other browser. You can search current builds or prior builds by gene name, genomic coordinates, accession numbers or keywords. When you run your search, you receive a Results summary, where you can view the search results in the context of gene, somatic mutation, transcript or variation. After you select your preferred context, you can access the browser view of your features. In the Gene view, you will see a list of transcripts and their translated protein products. In the Location view, you will see the genome browser view: your area of interest with surrounding features.

Use the “Configure this page” button to load different features on your genome map, like variation from dbSNP or HapMap, somatic mutations, regulation from histones, DNA methylation and much more.

The UCSC Genome Browser

The UCSC Genome Browser is the most dense and complex of the three genome browsers, but provides a wealth of genomic contextual information to users. Select your species and assembly from the drop-down menu on the home page and search by genomic position, gene symbol or search term. If you are searching by a gene with multiple transcripts, select the transcript version from your results and view the browser map. Above the map are buttons to zoom in or out; below the map is an array of menu options where you can see the default tracks loaded to the map and add more tracks from a vast selection of options.

The UCSC Genome Browser has comprehensive data from the Encode Project that can be loaded onto the maps, so you can view chromatin modification and other regulatory elements, expression data and lots more from the Encode project, as well as a wide variety of other experimental and predicted tracks.

To learn more about the UCSC Genome Browser, sign up for Tuesday’s class at the Galter Library on using the browser.

NCBI Map Viewer

The NCBI Map Viewer is possibly the simplest genome map of the three, but it has some unique features accessible from the viewer. You can access the MapViewer from the NCBI Gene record for any gene region, or access the MapViewer home page and select your species of choice, select a chromosome of interest, then zoom into a particular region of interest using the gene links or zoom ideogram on the left side of the page. You can add other maps to your view using the “Maps & Options” button, remove maps by clicking the X above any map, and access sequence and annotation data from the pink gene bar on the right.

MapViewer also has a feature called “ModelMaker”, accessible from the “Download/View Sequence/Evidence” window, in which you can build a sequence of putative exons from a gene to create a hypothetical transcript. The benefit of the NCBI MapViewer is that you can access genomic sequence faster than from any other browser.

Each of these browsers will display current genomic assemblies, genomic regions, exons, and transcription features, but each has its own set of strengths. Spend some time investigating each one and decide which is best for your purposes.

Monday's featured informatics resource: Galaxy.

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

Featured Informatics Resource 11-1-12: The Genetic Testing Registry

Thu, 01 Nov 2012 11:52:00 -0500

Thanks to reductions in the cost of sequencing individual genomes, individualized medicine and pharmacogenomics are promising to become future reality, rather than the stuff of science fiction. To assess and treat individual phenotypes of specific cancers or other inherited disease, genetic testing is more important than ever. That’s why today’s featured Informatics Week resource is:

The Genetic Testing Registry

The GTR is set to replace well-respected GeneTests in early 2013 as the primary source for locating genetic tests and testing facilities.

In April, I wrote a news article for this blog about the GTR replacing GeneTests, which is hosted by the NCBI, but is a product of the University of Washington. Since I wrote that article, the NCBI has been conducting numerous sessions for laboratories to migrate their data from GeneTests and learn to register their tests at GTR. GeneTests also manages GeneReviews, a high-quality series of expert-written reviews of inherited conditions. GeneReviews will continue to exist, and GTR includes links to GeneReviews and a search interface for the reviews.

Users can search for tests or facilities at GTR by

Tests – This is the default search at GTR. Search by test name, facility name, director’s name, gene symbol, condition, or disease
Conditions/Phenotypes – Search by disease name, traits, drugs
Genes – Search by gene symbol, name, condition name, or phenotype
Labs – Search by entering lab names, directors, staff members, locations and services, disease names, and phenotypes
Gene Reviews – search by any keyword

GTR has links to some useful and important resources in genetic counseling, such as directories of the National Society of Genetic Counselors, American Board of Medical Genetics and American Board of Genetic Counselors. GTR also links to OMIM, Orphanet and other variance resources.

The National Human Genome Research Institute (NHGRI) recently released a news article stating that the 1000 Genomes Project has more than doubled the number of known human genetic variants. While the NHGRI suspects many of these variants will likely produce no effect on health, some may be beneficial and some will likely contribute to disease. As these variants are discovered, sources of genetic testing will become ever more valuable. GTR helps to discover these tests and other valuable genetic testing and counseling resources.

Tomorrow's featured informatics resource: Genome Browsers

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

Featured Informatics Resource 10-31-12: GO Annotation Services DAVID & QuickGO

Wed, 31 Oct 2012 12:34:00 -0500

In yesterday’s blog post I wrote about GeneGo MetaCore, a pathway analysis software package that computes network interactions for a set of genes based on evidence from published literature and from Gene Ontology (GO) annotations. Today, I’m featuring two free web resources that rely primarily on GO annotations. Today’s featured resources:

DAVID and QuickGO

DAVID is a web service hosted by the National Cancer Institute. QuickGO is a web service offered by the European Bioinformatics Institute. They both allow a user to upload a list of molecule identifiers and find enriched GO annotations for that list, but they each have unique features that make them worth trying out.

Some users simply want lists of functional annotations associated with their experimental datasets. This is where DAVID is most valuable. DAVID’s Functional Annotation Tool accepts a list of object identifiers (gene symbols, Entrez IDs, Ensembl IDs, etc.) and performs enrichment analysis on this list of objects, returning lists of GO annotations sorted by the three main GO categories: biological process, molecular function or cellular component.

The Functional Annotation Tool goes beyond GO annotations to sort objects by many other taxonomies and biological ontologies: disease annotations, pathways such as KEGG and Biocarta; protein domains from SMART, PFAM and others; functional annotations from COG and more.

DAVID’s Functional Classification Tool calculates numbers of shared functional annotations between each possible pair of objects in a user’s list to find clusters of genes that may share functional characteristics based on shared annotations. DAVID also provides a very handy Gene ID Conversion Tool, which is very useful when you want to convert a list of identifiers (eg. Affymetrix SNP IDs, Agilent Chip IDs, RefSeq IDs) to corresponding identifiers from a different classification system.

QuickGO, like DAVID, will accept a list of object identifiers and sort them according to GO annotations. This is done by using the tools “Search and filter annotation sets” and “Investigate GOSlims”. QuickGO relies heavily on InterPro and UniProt Gene Ontology Annotation accessions and identifiers, so you may have to perform some extra steps to get it to work well if you are simply starting with a list of gene symbols. This can be done using QuickGO’s ID Mapping and Filter tools. QuickGO has a nice general search feature too, which allows users to simply type in a single term, like “apoptosis” or “breast cancer”, and it will retrieve all GO annotations associated with the term.

QuickGO has some extensive filtering capabilities, which allow the user to filter GO annotations based on experimental validation, electronic inference or inference based on sequence or structural similarity. This level of stringency control can help retrieve annotations that range in evidence validation from strict (direct experimental validation) to relaxed (electronic inference).

Related information:

With so much research being conducted on so many organisms’ genomes, the use of controlled vocabularies and ontologies is increasingly important. Ontologies describe not only terms, but their relationships to other terms.

If you’d like to learn more about:

Gene Ontology in general, visit the GO website
Other biological ontologies (and there are lots of them), visit the OBO Foundry website

Tomorrow's featured resource: the Genetic Testing Registry

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

Featured Informatics Resource 10-30-12: GeneGo MetaCore

Tue, 30 Oct 2012 12:43:00 -0500

Happy Chicago Informatics Week!

As I promised in yesterday's blog post, I will be featuring a different informatics resource every day during Chicago Informatics Week. Today's featured resource is:

GeneGo MetaCore

MetaCore is a pathway analysis software package from GeneGo, a company owned by Thomson Reuters (producers of Web of Science). MetaCore allows users to upload a gene list or dataset from an array experiment and perform enrichment analysis and other network interaction analyses on the data. MetaCore is the only resource I will be featuring this week that is NOT free and open source. However, the Galter Health Sciences Library currently pays the license for this resource so our users can benefit from its powerful network analysis functionality at no charge. To use the GeneGo MetaCore package, users must first register at the Galter website (details below).

GeneGo's search function can be used to search for a gene or protein to discover pathways in which that molecule is involved. The search functionality can also be used to find potential disease biomarkers by searching for a disease term.

The real value of MetaCore lies in its network building functionality. User datasets serve as "seed nodes" to build networks of interactions, based on functional interactions verified by published literature and Gene Ontology (GO) annotations. These interactions are curated by a team of scientists at GeneGo who read the scientific literature and build a knowledge base of pathway interactions that are used to build MetaCore's networks.

Networks can be built based on canonical pathways--known pathways such as apoptotic or oxidative stress--which are supplemented by GeneGo's curators. Networks can also be built based on transcription factor targets or protein functional interactions. Users can specify whether networks built are based only on direct interactions between their experimental molecules or are built with additional objects to create shortest paths of interaction between experimental objects.

If you're interested in using MetaCore and you don't already have access, read the information contained in our GeneGo pathways guide, including the terms of use, and click the link to register to use GeneGo MetaCore. You must sign in to your Galter Library account in order to register to use GeneGo. After you've registered, you will receive a username and password for access to the GeneGo user web portal, where you can log in directly and use MetaCore from any computer, anywhere. MetaCore users can access the GeneGo Tutorials & Training pages to view tutorials or contact me for assistance with the software.

Tomorrow's featured resources: Gene Ontology services DAVID (from the NCI) and QuickGO (from the EBI).

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

Galter Library and Chicago Informatics Week 2012

Mon, 29 Oct 2012 16:18:00 -0500

Day 1 - October 30, 2012 - Featured Informatics Resource: GeneGo MetaCore
Day 2 - October 31, 2012 - Featured Informatics Resources: GO Annotation Services DAVID and QuickGO
Day 3 - November 1, 2012 - Featured Informatics Resource: The Genetic Testing Registry
Day 4 - November 2, 2012 - Featured Informatics Resource: Genome Browsers
Day 7 - November 5, 2012 - Featured Informatics Resource: Galaxy
Day 8 - November 6, 2012 - Featured Informatics Resource: BioMart
Day 9 - November 7, 2012 - Featured Informatics Resource: FSM Informatics Projects and Research

Chicago mayor Rahm Emanuel has declared Oct. 30 - Nov. 7 2012 as Chicago Informatics Week. This announcement is scheduled to coincide with Chicago's hosting of the AMIA Annual Symposium, but will include much more than just the AMIA conference. Informatics Week is co-chaired by Justin Starren, Chief of the Division of Health and Biomedical Informatics at Northwestern University Feinberg School of Medicine, and Northwestern is hosting several related seminars and events. The Galter Library is getting in on the Informatics Week fun by posting special blog entries all week and offering a class on the UCSC Genome Browser.

To see a full listing of Informatics Week seminars hosted by Northwestern, University of Chicago, Rush University and UIC, check out the Events page for Chicago Informatics Week. Events include:

An informatics career session on the Evanston campus and an informatics "jam session" on the Chicago campus, both on Tuesday
Seminars, Wednesday through Friday at various locations, highlighting challenges and methods in informatics

Informatics is defined by the Merriam-Webster Dictionary as "information science: the collection, classification, storage, retrieval, and dissemination of recorded knowledge treated both as a pure and as an applied science". Extended to health and biomedical research, informatics has expanded to the fields of bioinformatics, medical informatics and biomedical informatics. Sometimes these terms are interchanged, but they all deal with using information in ways to solve complex problems of biology or medicine.

Since, at its heart, informatics is about management of information, the Galter Library is a perfect partner for informatics investigation. In honor of Informatics Week, I will be featuring a different bioinformatics resource each day on this news blog, beginning Tuesday October 30. Then, on Tuesday November 6, I will be offering a class on the use of the UCSC Genome Browser. To register for the class, go to the class information page.

So keep your eyes on this news blog over the next week to learn about some great informatics resources, and be sure to check out all of the great events for Chicago Informatics Week.

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

Datasets, Software Eligible for Listing in NSF Biosketches

Tue, 09 Oct 2012 16:23:00 -0500

This week, the National Science Foundation announced changes to its grant proposal guide (GPG). One of the changes is the acknowledgement of datasets, patents, software, and copyrights as citable products of research, eligible for inclusion in a researcher's biosketch. This is a great step forward in providing attribution to researchers' scientific output that falls outside of the traditional realm of manuscripts published in scholarly journals.

According to the GPG Summary of Changes, NSF biosketches have been revised to:

"rename the “Publications” section to “Products” and amend terminology and instructions accordingly. This change makes clear that products may include, but are not limited to, publications, data sets, software, patents, and copyrights."

Up to 10 total products may be included in the biosketch. Unacceptable products include unpublished papers and invited lectures.

Curation of databases, creation of datasets from large experiments, software for analysis of "Big Data", and patents are all important products of research, but don't effectively contribute to a researcher's h-index or other measure of impact when tenure and promotion boards review the researcher's curriculum vitae. Perhaps NSF's acknowledgement of these other intellectual products will spur more funding agencies (and the status quo of science scholarship in general) to look at impact as more than just numbers of published journal articles.

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

Do You Have Unpublished Figures or Data? Try figshare.

Wed, 12 Sep 2012 12:00:00 -0500

There's a new cloud data sharing platform available to researchers of all types. Created by a London PhD student, figshare (lowercase "f" intended) is a resource that allows researchers to share figures, datasets, null data and other research output that doesn't make its way into published manuscripts.

Probably the coolest new resource I learned about at this year's VIVO Conference, figshare provides users 1GB of private storage space, plus unlimited public uploads. figshare contains datasets, figures, posters, media and reports of null results. It contains files from many fields of research, from astronomy, physics and biological sciences to humanities. Files in the public space can be commented upon, tweeted and cited via downloads to Reference Manager, EndNote and Mendeley. figshare also provides counts of views, shares and will soon compile citation data on objects at the site.

Today's research produces so much material from supplemental figures to datasets that journal-published manuscripts often reflect only the tip of the iceberg of the volume of work done to produce publishable results. figshare provides a site to share these results with other researchers, but also provides a means of compiling non-traditional statistics on the impact of this scholarly output. When I recall all of the drawings and plots of brain sections that I created when I was a lab tech, I think a resource like figshare could have been very useful at my old labs. Materials shared on figshare stand a much better chance of reaching a wider audience of researchers than they do sitting in desk drawers or on computer hard drives.

If you have data, figures, media or other files that are not published and you want to share them, try figshare.

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

New Journal Links Datasets and Code to Articles

Fri, 27 Jul 2012 17:07:00 -0500

On July 12, a new peer-reviewed open access journal, titled GigaScience, published its first set of articles. GigaScience represents a novel format for science journals: manuscripts in GigaScience are published with links to software tools used for analysis and to corresponding datasets in the journal’s integrated database GigaDB.

The journal is a collaboration between BGI Shenzhen—the world’s largest genomics center—and the open access publisher, BioMed Central. The first issue of GigaScience includes a manuscript demonstrating the novel format of the journal. Stephan Beck’s group at the University College of London published an article on methods for whole genome analysis of DNA methylation. This article includes a section marked “Availability of supporting data”. This section links to an entry in the article’s references which displays a DOI link to the supporting data in GigaDB. The supporting data for this particular manuscript is 84 GB in size.

GigaDB will accept datasets up to 14 TB in size, so it is a valuable resource for authors who want to share their data or those who are required to do so by federal mandates such as the NIH Data Sharing Policy or the NSF Data Management Plan Requirements. The combination of manuscript, supporting data and links to software or code to analyze the data will improve reproducibility of computational results in experiments involving large datasets, and will also improve the likelihood of new discoveries from repurposed/shared data. The journal also falls in line with the White House Office of Science and Technology Policy goals on access to "Big Data".

The first issue of GigaScience also includes an editorial on GigaDB, a few commentaries, and an interesting review on the future of DNA sequence archiving, which had already been accessed over 2300 times within two weeks of its publication. This journal provides a valuable service, since many researchers may not have other means for online deposit of large datasets such as BAM files, FASTQ files and other large files produced by their experiments. Many other journals do provide access to supporting documents or supplemental files, but do not provide a means for storage of files of such a large size. It is my hope that the journal proves successful and sustainable, and that other journals consider duplicating this service.

For more information on GigaScience, see:

The BioMed Central blog post by the editors of GigaScience
The GigaScience journal blog, GigaBlog
Follow GigaScience on Twitter: @GigaScience

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

NSF Leads Federal "Big Data" Efforts

Fri, 30 Mar 2012 16:39:00 -0500

On March 29 2012, National Science Foundation (NSF) director Subra Suresh announced plans to expand NSF's funding, infrastructural, and scientific support of "big data" science. The announcement was made at an event held by the White House Office of Science and Technology Policy in Washington, D.C. The NSF was joined by the NIH and other federal agencies for the event.

Some people refer to this type of big data science as "e-science," though I've always preferred the term "big data science." Whichever label you use, it covers scientific disciplines from genomics to astronomy to geosciences and weather pattern modeling. Challenges to scientists involve storage, definition, and sharing of these massive datasets, and creation of statistical methods and algorithms for analyzing the data. The NSF requires all researchers applying for grants to submit Data Management Plans, and the NIH has a Data Sharing Policy: all investigator-initiated applications with direct costs greater than $500,000 in any single year are expected to address data sharing in their application or state why data sharing is not possible.

In his announcement, Suresh outlined details for a specific solicitation in joint partnership with the NIH for proposals for the program Core Techniques and Technologies for Advancing Big Data Science & Engineering (BIGDATA). This program "aims to advance the core scientific and technological means of managing, analyzing, visualizing and extracting useful information from large, diverse, distributed and heterogeneous data sets."

In addition to the BIGDATA solicitation, other programs in the Big Data effort include:

Cyberinfrastructure Framework for 21st Century Science and Engineering (CIF21)
CIF21 Track for IGERT
Data Citation
Data and Software Preservation for Open Science (DASPOS)
Digging into Data Challenge
EarthCube
Expeditions in Computing
Focused Research Group, stochastic network models
Ideas Lab
Information Integration and Informatics
The Computational and Data-enabled Science and Engineering (CDS&E) in Mathematical and Statistical Sciences (CDS&E-MSS)
Some Research Training Groups (RTG) and Mentoring through Critical Transition Points (MCTP) relate to big data
Laser Interferometer Gravitational Wave Observatory (LIGO)
The Open Science Grid (OSG)
The Theoretical and Computational Astrophysics Networks (TCAN)

To see brief descriptions of these programs, and to read details of Suresh's announcement, see the announcement page on the NSF website.

You can also find more information on federal big data initiatives at the White House Office of Science and Technology Policy website.

Related Galter Library Biosciences and Bioinformatics Blog Posts:

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

GTR: The Genetic Testing Registry

Mon, 12 Mar 2012 17:29:00 -0500

On February 29, 2012, the National Center for Biotechnology Information (NCBI) announced the release of the Genetic Testing Registry (GTR). The GTR is a resource that provides a "central location for voluntary submission of genetic test information by providers." The GTR contains information on genetic tests for approximately 2,500 diseases and will replace the GeneTests laboratory listings by 2013.

The GeneTests database, hosted by the NCBI but sponsored by the University of Washington, is a resource long known and used by genetic counselors and other genetic testing and healthcare professionals. The GTR strives to continue GeneTests' service of providing information supplied by gene testing organizations. The GTR, while replacing GeneTests, will not replace GeneReviews, expert-authored reviews on genetic conditions, and in fact includes links to GeneReviews.

Users can search the entire GTR site using keywords, or can search the GTR by specific categories:

Tests - by entering test names, disease names, phenotypes, gene symbols and names, protein names, laboratory names, directors and locations as search terms
Conditions/Phenotypes - by entering disease names, traits, drugs, proteins and analytes
Genes - by entering gene symbols and names, conditions and phenotypes
Labs - by entering lab names, directors, staff members, locations and services, disease names and phenotypes
GeneReviews - by entering any keyword

Search results can be narrowed by applying filters selected from the left side of the search results page. The GTR homepage also includes links to a number of other genetic testing resources. Researchers or laboratories who provide genetic testing can also find links and instructions on submitting genetic testing information to the GTR.

The GTR is somwhat of a hybrid resource. It is intended for use mostly by researchers and clinicians. It does contain information that can be used by patients and consumers, but also includes advice to consumers to contact their healthcare providers or genetics professionals with questions about tests.

For more information about the GTR, see the GTR FAQs from the Office of Biotechnology Activities.

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

Previewing the 2012 NAR Database Issue

Tue, 06 Dec 2011 17:05:00 -0500

If you're like me, you look forward to the annual release of the Nucleic Acids Research (NAR) Database Issue and the Web Server Issue. The NAR 2012 Database Issue is due in January, but several papers are available in advance as early open access releases. The new issue promises 92 new databases and updates to some of your favorite essential resources.

The 19th annual Database Issue features papers on:

neXtProt - a knowledgebase of human proteins
Papers on the BioSample databases from the NCBI and the EBI
Updates to Gene Ontology (GO) and the UCSC Genome Browser
New and updated small molecule databases
New Wiki-style annotation formats for databases
Updates from the International Nucleotide Sequence Database Collaboration (INSDC)
Papers on the commercialization of the Kyoto Encyclopedia of Genes and Genomes (KEGG), which was formerly free to access; and the possible termination (or commercialization) of the Arabidopsis Information Resource (TAIR)
A paper by INSDC members on the sustainability of genome databases, with some discussion of the Sequence Read Archive

I'm especially excited by the BioSample databases, relatively recent launches from the NCBI and EBI. In this time of expansive research in genomic medicine and subsequent increase in DNA sample collection, the BioSample databases provide much-needed centralization of descriptions of biologicial samples worldwide. I also enthusiastically read about the updates to some of my favorites (UCSC Genome Browser and GO), but I am dismayed by the uncertain future of KEGG (read the plea to support KEGG by founder Minoru Kanehisa). This issue highlights the promises of and the threats to the future of funding, support and hosting of biomolecule databases.

For highlights of the upcoming NAR 2012 Databases Issue, see the advance access version of the summary paper by Michael Y. Galperin of the NCBI and Xose M. Fernandez-Suarez of EMBL-EBI.

You can also see 100 advance access papers from this issue at NAR's Selected Abstracts page for the 2012 Database Issue.

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.

Revisiting "Bioinformatics: Gone in 2012"

Tue, 06 Dec 2011 13:06:00 -0500

In 2003, at the O'Reilly Bioinformatics Technology Conference, Lincoln Stein delivered a keynote address titled "Bioinformatics: Gone in 2012." With 2012 just around the corner, let's revisit Stein's speech and compare his predictions with the state of bioinformatics today.

Stein, a researcher at Cold Spring Harbor Laboratory, made several statements in his keynote address (summarized here by Daniel H. Steinberg) concerning the nature of bioinformatics as a discipline. His main point was that bioinformatics "won't be a field in 2012," though researchers working in bioinformatics would still have plenty of work. It may just be a question of semantics and how we define bioinformatics, but was Stein right or wrong? The answer to this question is yes, he was both right and wrong. Let's look at some of his points:

What is bioinformatics?
This particular topic in his speech is amusing for me, because I still hear researchers debating what bioinformatics is. Everyone has a slightly different definition, so many people refer to bioinformatics.org's useful FAQ and definitions of bioinformatics. Stein referred to other researchers' definitions of bioinformatics as 1) DNA and protein analysis or 2) managing large data sets. Both definitions still apply today.

But is bioinformatics a field?
Here, I will have to respectfully disagree with Stein's opinion that bioinformatics would not be a field by 2012. He argued that bioinformatics was a tool (or set of tools) like microscopy, used to answer complex biological questions. I would agree with that, but I also think that bioinformatics qualifies as a field as well. Perhaps I need to justify my own graduate degree in computational biology and bioinformatics, but I prefer to look at bioinformatics as an integrative discipline, not just a set of tools. Read on to see why...

The "department of" test
Stein posited that bioinformatics does not stand up to the "department of" test. Meaning: take the name of the field, and put the words "department of" in front of it.

Stein compared bioinformatics to microscopy: both are techniques or tools, but do not stand alone as disciplines. Most universities don't have a Department of Microscopy and Stein predicted that a Department of Bioinformatics was equally unlikely. Do departments of bioinformatics exist in universities in 2012? The answer is "yes, they do" (Emory, Bioinformatics and Genomics at UNCC). Many more universities have divisions of bioinformatics within other departments, so Stein may be partly correct about bioinformatics not standing up to the "department of" test.

I think that bioinformatics is more appropriately compared to statistics or biostatistics rather than microscopy: one can argue that statistics is a tool for analysis, but statistics as a field is also strong enough to merit departmental status. So is bioinformatics. In fact, some universities have "Departments of Biostatistics and Bioinformatics" (Duke, Georgetown). These two disciplines cut across other fields of investigation, but are practiced by their own trained experts. At Northwestern, we don't have a standalone "Department of Neuroscience," but that doesn't mean that neuroscience isn't a field. It's an interdepartmental field of study. So is bioinformatics. Many universities have graduate programs in bioinformatics without having departments of bioinformatics, so I think that here, Stein was right and wrong: bioinformatics is both a field and a set of skills.

The right experts for the job
Many people can use bioinformatics tools on their own without having to consult a bioinformaticist, just as many researchers can do their own statistics without consulting a statistician. Bioinformatics is a field with so much complexity of both biology and computational expertise that most institutions, Northwestern included, have bioinformatics centers or core facilities (like Northwestern University Biomedical Informatics Center (NUBIC) here at the Feinberg School of Medicine). Librarians, in many ways, are like statisticians or bioinformaticians: we use tools that many people can use on their own. We all do work that cuts across disciplinary boundaries and we employ strategies that many users may not be aware of. Our realms of expertise are familiar to many, but expertly practiced by few.

Is it just semantics? Is there a future?
So is bioinformatics a field? Yes. A discipline? Yes. A set of tools? Yes. There are conferences of bioinformatics; divisions, centers, departments and graduate programs in bioinformatics. It's fun to go back and read the summary of Stein's 2003 keynote address. Genomic and translational medicine have created a large demand for skilled bioinformaticists who are comfortable with both the biology and the computational demands of genome analysis, pathway analysis and molecular modeling. Stein was right about bioinformatics being a set of tools, but it seems that, as a discipline and field of expertise, bioinformatics is still very much here and not likely to be "Gone by 2012".

Pamela Shaw
Biosciences & Bioinformatics Librarian
email Pamela

The Biosciences & Bioinformatics Blog highlights new tools and news items of interest to the biosciences research community at Northwestern University.