YouTube Facebook LinkedIn Google+ Twitter Xingrss  

Gene Browser with a Google Feel


By Laurie Wiegler

March 1, 2008 | Following the launch last year of ECell 3D, Kazaharu Arakawa of Keio University in Japan has launched another colorful and intuitive bioinformatics tool. Drawing inspiration from iTunes and Google Maps, the database browser, called Genome Projector (www.g-language.org/GenomeProjector/), was developed at Keio’s Institute for Advanced Biosciences over the past year.

As a highly visual and searchable genome browser with zoomable user interface, Genome Projector contains four “maps:” Genome Map, Plasmid Map, DNA Walk, and Pathway Map. Each map indicates genes in various colors, such as rRNA in red stripes and tRNA in green stripes in the Genome Map.

Genome Projector currently lists some 320 circular bacterial genomes, reflecting Arakawa’s research interests, but may be expanded to include eukaryotes.

Design was important to Arakawa. He’s a fan of Apple software because it traditionally sports an intuitive user interface. “This is often not the case for bioinformatics software and databases,” he says. “[I] borrowed color schemes and layout [from iTunes],” Arakawa says, adding that he was also inspired by Google Maps, from which he plucked its free API for the project design.

His inspiration, however, was a Roche biochemical pathways chart. As an undergraduate, he was fascinated by the metabolism wallchart. It seemed to him “the grand sum of the history of biochemistry,” he recalls. He “almost trembled with excitement to start afresh in systems biology, to build dynamic cellular simulations on top of this knowledge.”

The result of that fervor is Genome Projector. The G-language-powered database browser uses a Restauro-G back end to annotate the genome by similarity searches using the BLAST-Like Alignment Tool, connecting with protein databases such as UniProt Knowledgebase and NCBI nr. The front end boasts Web 2.0 technologies such as AJAX and EXT 2.0 framework.

Distinguishing Itself
Genome Projector resembles Cold Spring Harbor researcher Lincoln Stein’s GBrowse, although the two browsers do differ. For example, Arakawa says browsers such as GBrowse are designed to be scalable interfaces but require several user interactions, and are not genuinely zoomable user interface (ZUI) applications.

Stein told Bio•IT World: “That is absolutely correct! However, there is a trade off between speed and functionality, and the popular browsers — I’m including Ensembl and UCSC genome browser in this — provide more functionality at the moment.”

Stein says features need to be added to Genome Projector to broaden its appeal, such as a way to allow for annotation, so “community members can upload data sets to visualize... the genome background,” and a way to show more detail as one zooms in, via semantic zooming. While, “it does have popup windows with additional information on genes and enzymes… the lack of semantic zooming is still a big usability problem, in my opinion,” says Stein.

Arakawa says his primary aim was to “allow quick, continuous zooming as opposed to the original KEGG [Kyoto Encyclopedia of Genes and Genomes] Pathway and GBrowse, [which both] require page transition[s] for zooming in and out.” But he concedes that if his team implements a human map for Genome Projector, “it is critical to consider using semantic continuous zooming.”

Other critics charge that Genome Projector may not be viable for large-scale genomic analysis. Accelrys’ Nancy Latimer, product manager for the Gene Expression, R-Statistics, and Text Analysis collections for the company’s SciTegic Pipeline Pilot products, is doubtful.

“From an information/visualization point of view, it is extremely compelling; it’s very, very intuitive and it satisfies a lot of the requirements of a good data visualization approach,” she says. However, she is “less certain” that for researchers doing genomic analysis, “this tool would give them the flexibility they need to zoom in on their particular question.

Mapping the Next Steps
Genome Projector is slated to go beyond beta by the end of the year, after publication, but Arakawa is already busy with plans for its next phase.

He hopes to introduce more views, starting with KEGG Atlas. “Right now, the only way to map certain information over the four views is by creating a new overlay, which is computationally heavy [and takes] several minutes to generate,” Arakawa says. “I would like to optimize this process so that we can start providing a web service for data mapping, and to develop a new means based only on client-side Javascript so that it is faster with no load on the server.”

Moreover, he would like to make it easier for Genome Projector users to put their own pins and annotations on the maps, to allow sharing of information, ideally with a DAS (Distributed Annotation System) backend.

Genome Projector’s future may include a resemblance to the “My Maps” feature of Google Maps.

Sidebar: ECell 3D

Arakawa debuted ECell 3D at a packed house at an Intelligent Systems in Molecular Biology conference workshop last July in Vienna. Like Genome Projector, ECell 3D is a bioinformatics tool with a focus on ease of use.

“In my experience, [the] modeling process is mostly heuristic, with numerous trials and errors to get the model to work right, to sufficiently capture the cellular behavior that a scientist wants to test and learn,” says Arakawa. Therefore, good user interface and design that supports this process, such as ECell 3D for simulation results, and Genome Projector for data mining and searching, are essential.”

ECell 3D boasts an interface that enables researchers to model cellular systems in 3-D graphics. The resulting platform, controlled by Bluetooth, is strikingly visual, looking much like a flight pattern or constellation map.

Arakawa hopes to start integrating the software tools he’s developed so far to create a prototype modeling environment around ECell and hopes to achieve a “sophisticated environment” for modeling in systems biology.

“There are a number of cell simulators already available,” says Arakawa, but he believes “Automation through knowledge integration for modeling is a key step for the success of systems biology.”  -- L.W.

 

________________________________________________

 This article appeared in Bio-IT World Magazine.
Subscriptions are free for qualifying individuals. 
 Apply Today.

View Next Related Story
Click here to login and leave a comment.  

0 Comments

Add Comment

Text Only 2000 character limit

Page 1 of 1


For reprints and/or copyright permission, please contact  Terry Manning, 781.972.1349 , tmanning@healthtech.com.