The second sbv IMPROVER Challenge was in many ways more complex than the first, with more data to consider, in rather challenging settings. We were glad to see that the approach we refined for the previous challenge proved once more to be robust. Once again, sbv IMPROVER has proven itself to be unique in allowing us to identify which predictive techniques work best in specific, distinct settings, and at the same time giving us a perspective on the limitations of rodent models in the study of human biology.

Dr. Adi Laurentiu Tarca
Director, Bioinformatics and Computational Biology Unit, Perinatology Research Branch (NCIHD/NIH).
On the Species Translation Challenge

The results from each sub-Challenge may have far reaching implications for scientists in many fields that use animal models to understand more about human biological systems. It is clear that from a computational biologists’ perspective, rodents and humans are indeed closer than we think. We have been delighted with the quality of submissions to the Species Translation Challenge. If the goal is to make meaningful predictions about human biology from data derived from rodent models, it is extremely important to carefully consider the specific characteristics of the experimental system and the biological mechanisms that may be impacted in order to identify the most adequate model systems for a given question.

Prof. Manuel C. Peitsch
VP, Biological Systems Research, Philip Morris International R&D.
On the Species Translation Challenge

Among the issues in our understanding of species translatability is the need to bridge the gap between informatics and real-world biology. sbv IMPROVER continues to provide valuable insights in this area, bringing together scientists that understand innovative techniques such as machine learning and neural network with molecular and cell biologists. In doing so the project addresses both the technical and biological challenges inherent in making predictions between species.

Prof. George Kollias
Biomedical Sciences Research Center ‘Alexander Fleming’, Vari, Greece.
On the Species Translation Challenge

It is clearly very difficult to make direct predictions of human biology based on rodent models. However, we can see that relationships that hold within rodent gene-sets, identified through our computations, are themselves useful in helping us understand what happens in humans.

Dr. Sahand Hormoz
Postdoctoral Fellow, Kavli Institute for Theoretical Physics, University of California Santa Barbara.
On the Species Translation Challenge

With a diminishing return on investment on drug development, translational modeling is becoming an increasingly important focus in the pharmaceutical industry. The ultimate power lies in being able to confidently verify translational concepts in order to build truly informative clinical models. The techniques used in the sbv IMPROVER Species Translation Challenge, both in terms of the submissions and the challenge design, are highly relevant in helping us achieve this goal.

Dr. Carolyn Cho
Director, Modeling & Simulation, Global Development, Merck Sharp & Dohme (MSD).
On the Species Translation Challenge

The Species Translation Challenge invited us to ask: are rats and humans closer than we think? Based on our work on the second sub-Challenge, it seems that there may well be a closer relationship between rat and human than what is currently acknowledged. This is an exciting discovery, with potential implications across a number of fields, and one we look forward to exploring further.

Prof. Michael Biehl
Johann Bernoulli Institute for Mathematics and Computer Science, University of Groningen, Netherlands.
On the Species Translation Challenge

The sbv IMPROVER Network Verification Challenge brought together scientists from around the world in many different disciplines to share their expertise in biological signaling. This allowed for specific feedback to refine the networks that does not exist in traditional scientific forums. Being able to have an active discussion with scientists who have studied these pathways their whole career was a stimulating experience that will enable us to more clearly and comprehensively describe the biology in these networks. Having both keynote speakers in a traditional speaker-audience format along with breakout sessions where everyone could voice their opinions made for a unique environment that allowed for a high amount of contribution from all participants at the meeting.

Dr. Jennifer Park
Senior Scientist, Selventa, USA.
On the Network Verification Challenge Jamboree 2014

Under the umbrella of the sbv IMPROVER Network Verification Challenge, scientists from around the world are working together to build the most comprehensive and sophisticated models of COPDs that we have ever known. These models will be invaluable in assessing the mechanistic action of new drug compounds, and thereby expedite the drug discovery and development process. Beyond COPD, many of the models are relevant to other diseases too, and the implications for the advancement of healthcare across a number of fields are thus far reaching.

Dr. Daniele Guardavaccaro
Principle Investigator, Hubrecht Institute, Netherlands.
On the Network Verification Challenge

Building accurate biological maps is a complex process that requires continuous review to ensure all new discoveries are appropriately captured. The unique design of the sbv IMPROVER Network Verification Challenge allows us to build maps of cross-interacting factors (mediators, enzymes, receptors, signaling molecules etc), thereby helping us to understand basal and pathologic mechanisms as best as possible, and potentially to identify new pharmacological targets to treat disease.

Dr. Maria Laura Belladonna
Department of Experimental Medicine, University of Perugia, Italy.
On the Network Verification Challenge

The Network Verification Challenge demonstrates just how effective crowd-sourcing can be to build, test and validate complex biological models. We have used the outputs of the first stage of the Network Verification Challenge as the gold-standard for our own text mining evaluation campaign, which is being run as part of BioCreAtIvE (Critical Assessment of Information Extraction systems in Biology). With the Network Verification Challenge now attracting many more participants we are seeing an even greater level of granularity emerge. The robustness of the methodology, together with the high level of activity we are seeing in the challenge, means that the models can be used with confidence in a whole range of applications.

Dr. Fabio Rinaldi
Institute of Computational Linguistics, University of Zurich, Switzerland
On the Network Verification Challenge

It is great to see PMI taking a network biology approach to the study of smoking-related diseases. As well as being useful in the biological context, the models produced by initiatives such as the Network Verification Challenge also have important regulatory implications, providing a verifiable framework for data analysis and improving the human-relevance of regulatory test methods.

Dr. Kristie Sullivan
Physicians Committee for Responsible Medicine, USA
On the Network Verification Challenge

A key design goal of the Biological Expression Language is that the language be easy to learn and understandable to lab scientists, not just bioinformaticians. It is fantastic to see it used for the Network Verification Challenge, showcasing the usability of the language and expanding the OpenBEL community.

Dr. Natalie Catlett
Selventa, USA
On the Network Verification Challenge

The Network Verification Challenge presents an opportunity to redefine COPD to the benefit of real-world clinical practice. The models being produced have the potential to complement clinical experience with a robust, molecular-level understanding of COPD, which ultimately can translate to improved disease diagnosis and patient management.

Dr. Michael Liebman
IPQ Analytics, LLC, USA
On the Network Verification Challenge

sbv IMPROVER is shaping up to be the new gold-standard in assessing scientific validity. The exponential growth of scientific data presents us with a twofold problem: how can we generate models that are both correct and complete? sbv IMPROVER addresses this by using the power of the people, through its crowd-sourcing challenges, to give us confidence in the accuracy of our findings.

Ashutosh Malhotra, M.Sc.
Researcher, Department of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI), Germany; PhD Student, Bonn-Aachen International Center for Informatics, University of Bonn, Germany.
On the Network Verification Challenge

The techniques being used in sbv IMPROVER are the future of systems biology. From the novel models being developed in the Network Verification Challenge we will be able to develop robust hypotheses of gene expression changes and pathway activation/quiescence. For the first time, we can predict the potential therapeutic impact of retinoids on COPD with a degree of precision and accuracy never previously realized.

Dr. David Kling
Visiting Scholar, Department of Biology, Boston College, USA.
On the Network Verification Challenge

The first sbv IMPROVER challenge offered an excellent opportunity to test and develop different approaches to classifying clinical samples based on gene expression. We were hopeful that the challenge would inform our work in the prediction of adverse pregnancy outcomes and the results are looking very promising. Our analytical approach was designed to be robust enough to handle the different sources of non-biological variability in the data, while identifying small sets of biomarkers predictive for each of the diseases. Unlike most of the top ranked teams, two of the four models we proposed used information from as little as two genes instead of hundreds.

Dr. Adi Laurentiu Tarca
Assistant Professor (Research), Department of Computer Science & Center for Molecular Medicine and Genetics, Wayne State University.
On the Diagnostic Signature Challenge

It has been wonderful to see how sbv IMPROVER has captured the attention of the scientific community. The volume and quality of the submissions to the first sbv IMPROVER challenge has been truly inspiring. It has been a pleasure to invite scientists and academics from around the world to use this state-of-the art approach to methods verification and help further our understanding of concepts in systems biology. I believe we are off to a great start for the sbv IMPROVER project and we hope to see many more exciting developments in the months and years to come.

Prof. Manuel C. Peitsch
VP, Biological Systems Research, Philip Morris International (PMI), Research & Development.
On the Diagnostic Signature Challenge

As the cost of sequencing human genomes drops to $1000 and below, our capacity to generate data is far outstripping our ability to interpret it. Projects like sbv IMPROVER are important because they challenge scientists to identify systems biology methods that are best able to make meaningful predictions and, hopefully, improve outcome for patients.

Prof. John Quackenbush
Professor of Computational Biology & Bioinformatics, Dana-Farber Cancer Institute.

The sbv IMPROVER project has the potential to lead to huge strides forward in our understanding of biological systems. Advanced technologies such as next generation sequencing and high-resolution mass spectrometry based proteomics are producing rich and complex datasets. By applying the computational power of the sbv IMPROVER community to the outputs of these technologies, we should be able to gain deep, robust and highly predictive insights into disease progression and therapeutic intervention.

Dr. Martin Hornshaw
Thermo Fisher Scientific (UK)

We need new systems to effectively and usefully deal with the complexity of high-throughput data. sbv IMPROVER represents one such system with a number of potential applications, from identifying predictive biomarkers of disease to using network analysis techniques to help understand drug action mechanisms.

Dr. Joaquín Dopazo
Head of the Computational Genomics Department, Principe Felipe Research Center.

sbv IMPROVER has given us the opportunity to put together an interdisciplinary team and use tailored computational techniques on this unique, exceptionally high-quality data-set. We are delighted our approach has proved so successful and that we have been able to contribute to this important initiative on species translatability. It is crucial that we continue to take steps to improve modeling capabilities as a complement to in vivo systems, and we would like to see how we could improve our predictions now that the Challenge is closed and the full data-set has been unblinded.

Prof. Gyan Bhanot
Rutgers University.

The sbv IMPROVER project is leading the way in bringing together multiple scientific disciplines to look at how different systems interact with each other within the human body. As we move towards a more holistic approach to healthcare, where the goal is to empower people to maintain health rather than treat isolated manifestations of disease when they occur, the implications of sbv IMPROVER will be relevant not just to this diverse set of scientists, but ultimately to the global public at large.

Prof. Bruno Sobral
Assistant VP & Head of BioSystem Informatics, Nestlé Institute of Health Sciences, Switzerland.

sbv IMPROVER continues to grow in size and complexity and in doing so increases the possibility of identifying new pathways in health and disease that will certainly lead to the development of more efficacious drugs and treatment regimens. I salute all that are involved and encourage more to join so that the scientific community can realize the full value of the system.

Dr. Sam Hou
SVP, Head Biologics Research, Glenmark Pharmaceuticals SA, Switzerland.

We were driven by the desire to create a model that can both lead to valuable biological insight, and be implemented in practice at the lowest possible cost. The Systems Toxicology Computational Challenge has allowed us to test the quality of our research and I’m delighted that our approach has proved to be robust.

Prof. Adi Tarca
Wayne State University, School of Medicine

sbv IMPROVER is a great example of how a crowdsourced challenge can be used to provide both transparency and confirmation to the diagnostic development process. The fact that the challenge winners used different analytical techniques but arrived at a similar core gene signature for smoking response represents an important validation milestone in preparation for proceeding with product development.

Dr Bob Terbrueggen

Assessing toxicant exposure from blood data alone is a complex and fascinating undertaking. The tasks set in the Systems Toxicology Computational Challenge allowed us to address this issue, while objectively testing our computational methodologies. As computational biologists we are always looking for new data to work with, and it is of huge value to have access to the high quality datasets provided by sbv IMPROVER.

İsmail Bilgen
Istanbul Technical University

sbv IMPROVER is utilizing the knowledge of the crowd to address specific biological problems. In classifying smokers and non-smokers using gene signature data, we needed a model that could work on independent datasets. To achieve this we applied open source machine learning tools for feature selection and prediction, optimising our algorithm with a cross validation approach.

Dr Sandeep Kumar Dhanda
Division of Vaccine Discovery, La Jolla Institute for Allergy and Immunology

Share this page