Nature Network - Recent topics from Integrative Bioinformatics & Network Biology

Computational biology resources: lack of persistence and usability? (1 reply)

Erfan Younesi — Tue, 22 Jul 2008 09:37:24 -0000

This is a long-standing problem in the field of computational biology and I’m sure almost every user has experienced such scenario: you follow a link from a journal article (which has been published 2-3 years backwards) to an online database or software; either the webpage does not exist anymore or the tool is not properly functional! Certainly you feel frustrated specially when you’re urgently in need of that functionality for your research. But why is it so?
There are hundreds, if not thousands, software programs and data repositories which have been developed for specific research topics or applicability but have been left without being extended to novel applications or being updated. In the era of integrative bioinformatics and systems biology, one needs to integrate data from existing resources efficiently; thus any inconsistency in the content as well as performance of these resources could lead to misleading results.
Philip E. Bourne and his colleagues from University of California San Diego, La Jolla, have brought this issue forward in their commentary at PLoS Computational Biology entitled Computational Biology Resources Lack Persistence and Usability.
At the end, they ask: “Wouldn’t it seem that evidence of usability through suitable documentation and accessibility should be prerequisite to publishing a paper when that paper is about such a resource? Is our integrity being compromised by the resources we are making public?”
Personally, I think the solution of this problem could be quality control, monitoring, and maintenance of bioinformatics resources by a third party organization.
What do you think?

microRNAs: New Players in Network Biology (1 reply)

Erfan Younesi — Thu, 08 May 2008 21:02:13 -0000

Today I was among the audience of Prof. John Quackenbush, the well known scientist in microarray data analysis from Harvard School of Public Health. He presented a network approach his team has adopted to build up a directional interaction network of genes by integrating gene expression variations into a literature-based network of protein-protein interactions. The interesting point in his presentation was integration of an additional layer of information, namely RNAi knock-out array.
His answer to my question about existence of public repositories for accessing RNAi array data was frustrating:”No such a public database exists”. Although profiling microRNA gene expression has not got that much popularity yet, thinking of storing these expression data in a publicly available database can motivate systems biologists to add an additional “informative” dimensionality to their predictive models.

Predicting Missing Connections in Networks: A New Support for Systems Biology? (0 replies)

Erfan Younesi — Tue, 06 May 2008 22:29:27 -0000

In the first week of this month (1 May 2008), an article appeared in Nature (Volume 453, Number 7191) that can be viewed as an important step toward unraveling the real structural properties of networks, including the biological ones. In their publication, the researchers in Santa Fe Institute suggest that hierarchy is a central organizing principle of complex networks which can be used to discover missing relationships between entities. Probably the most interesting part of their work is development of novel algorithms (based on the hierarchical organization) which are able to “predict” the missing links from partial data with high accuracy. Even when only half of the connections in three networks were shown to their algorithm, the researchers found that hierarchical structure can predict missing connections with an accuracy of up to 80 percent.
Although their algorithm worked well on 3 types of network (the metabolic network of the spirochete Treponema pallidum, a network of associations between terrorists, and a food web of grassland species), it would be interesting to see how this approach can contribute to increasing our understanding of the systems biology of human complex diseases, including cancer.

MIMIx: A New Solution for an Old Problem (0 replies)

Erfan Younesi — Mon, 19 Nov 2007 22:58:40 -0000

Those of you who have worked with molecular interaction databases (specially the protein-protein interaction data) know very well that heterogeneity and ambiguity of interaction data in majority of databases are two big hurdles to researchers. In order to overcome this problem, several efforts have been made including that of Oliva et al. who introduced PIANA approach for protein interactions and network analysis. Such approaches use a different parser for each database in order to transform the data into a common format.
In a paper published in Nature Biotechnology on 8 August 2007, Sandra Orchard et al. have reported a new standard namely “the minimum information for reporting a molecular interaction experiment” which proposes a guideline for bench scientists to avoid inconsistencies and missing information in their published papers. The aim of MIMIx is to follow the PSI-MI standards and to support the systematic capture of molecular interaction data in public databases.

RNA silencing networks (0 replies)

Erfan Younesi — Sat, 03 Nov 2007 23:30:28 -0000

During the “23rd Ernst Klenk Symposium on Non-Coding RNA” in 30 Sep. 2007 which was held in the Medical Faculty, University of Cologne, I was fascinated by a talk about the RNA silencing networks in plants.
In this talk, Professor David Baulcombe from the Sainsbury Laboratory, UK, explained how siRNAs can be amplified and stimulate further rounds of siRNA production in the model plant, Arabidopsis thaliana. This amplification apparently triggers a cascade or network of gene silencing.
Here some interesting questions arise: How does this regulatory mechanism perform in concert with signaling pathways? Is this cascade limited to the similar genes involved in the same pathway? What are the properties of silencing networks?

RNA silencing networks (0 replies)

Erfan Younesi — Sat, 03 Nov 2007 23:29:28 -0000

During the “23rd Ernst Klenk Symposium on Non-Coding RNA” in 30 Sep. 2007 which was held in the Medical Faculty, University of Cologne, I was fascinated by a talk about the RNA silencing networks in plants.
In this talk, Professor David Baulcombe from the Sainsbury Laboratory, UK, explained how siRNAs can be amplified and stimulate further rounds of siRNA production. This amplification apparently triggers a cascade or network of gene silencing.
Here some interesting questions arise: How does this regulatory mechanism perform in concert with signaling pathways? Is this cascade limited to the similar genes involved in the same pathway? What are the properties of silencing networks?