Plectix Blog

Neuroscience and Systems Biology

By Vy Hoang

With recent advances in genetic and functional genomic methods, the field of neuroscience is playing catch up with the other areas of biology that have already adopted these techniques (Geschwind et al.). Some reasons behind its slow start are related to the complexity of the neural circuits, the dependency on post-mortem human tissue, and the massive amounts of data generated from neuroscience laboratories. However, these new technologies will also generate lots of data across multiple levels (phenotypic, genotypic, epigenetic, proteomic, and genomic). The question now is how do neuroscientists integrate these different layers of data? The answer is that they largely haven’t!

Using a reductionist view, neuroscientists have traditionally divided the brain into anatomical regions and have then tended to study each region in detail, but also in isolation (Villosalda et al.).  Although this approach has had its successes, the next step to understanding the complexity of the brain as a whole with regard to functions like memory, learning, and behavior is to integrate the knowledge of the different anatomical regions. Naturally such an ambitious goal is easier said than done.  Some sort of standardization in the protocols used for neuroscience experiments might be a step in the right direction, allowing researchers to combine experimental data more readily.

Neuroscientists might do well to consider the systems biology approach in which the components of a system are considered at multiple levels of detail whose properties are interrelated (Villosalda et al.). By enabling neuroscientists to integrate data in ways that allow them to study the brain as a whole, systems biology may help them to understand the mechanisms underlying neurological diseases such as Parkinson’s disease and schizophrenia since their etiologies involve more than one section of the brain. The elucidation of these disease mechanisms would enable medical researchers to more clearly define neurological targets of interest and to develop safer and more effective therapies. Although the field of neuroscience may currently be lagging behind in embracing the systems biology approach, it seems likely that its researchers will also seek to take advantage of the benefits that it can offer before too much longer.

The Death of Skepticism

By Isha Antani

In recent times, systems biology has been “rediscovered”. Systems biology has been around for a while, often referred to by different names. Papers were published as early as 1952 that explored the concepts of modeling cellular systems (e.g., Hodgkins and Huxley’s Nobel Prize winning paper on a mathematical model of the action potential propagating through an axon). Until recently, however, it was never much in the mainstream and scientists, for the most part, continued to focus more on specific molecular biology or on looking at biology from a larger, cellular perspective but never forging a link between the two. Since the turn of the 21st century, this area of science has been slowly gaining steam. The International Conference on Systems Biology (ICSB) was started in 2000 and each annual conference has been growing in number of attendees ever since. It was only a matter of time before the pharmaceutical industry recognized the potency of the field. However, as quickly as the pharma companies jumped on board, they hopped off board when the science failed to deliver fast results.  

Now given some room in which to breathe, the field of systems biology has gotten time to flesh out and develop. A multidisciplinary field, merging computer science, biology, and engineering, it has now had the needed time to create a strong message and a thriving community. In the last few years, technology catering towards this field has improved considerably and has led to the acquisition of and better digestion of large, meaningful data sets about protein-protein interactions. Additionally, as better-trained scientists emerge from the multitudes of recently created Systems and Computational Biology university programs, the field has started to generate tangible results from model-driven experiments.

For starters, Cambridge-based Merrimack Pharmaceuticals has successfully developed the first drug ever motivated by a systems biology model, MM-121. MM-121, an ErbB3 inhibitor, entered phase 1 in late 2008. In June 2009, another Merrimack drug MM-111, which simultaneously targets ErbB2 and ErbB3, entered clinical development in a Phase 1/2 trial. 

Pharmaceutical companies such as Roche that tentatively invested into the field have also started seeing effective results and quicker than expected. In 2005, Roche collaborated with SystemsX, a Swiss Initiative in Systems Biology, on a project titled Systems biology of the beta cell- application to type 2 diabetes progression. There was some skepticism about what this partnership would be able to accomplish in the three-year run the partnership was initially given. However, within a year of the project start, the group headed by Cristiano Migliorini was able to drill down to a single, clear target. The identity of the target protein is still kept secret. 

As additional systems biology-derived drugs enter the pipeline in the next few years and eventually make it into the marketplace, then perhaps in the next few years skepticism in the pharma industry will diminish. Surely enough, a recent report by PricewaterhouseCoopers stresses the need for the incorporation of in silico methodologies in lab regimens and predicts that “by 2020, virtual cells, organs, and animals will be widely employed in pharmaceutical research, reducing the need to experiment on living creatures.”

What's New In Cellucidate?

By Traci Freed

Exciting new functionality, including collaboration tools, has recently been implemented in Cellucidate.  Because Cellucidate is web-based there are no updates to install on your computer so login to your account and start exploring these features.  If you are not a registered user, you can sign up at www.cellucidate.com.

Collaboration Tools
Cellucidate has made it easy to collaborate with other users.  Just add users as collaborators to your cBook, choose which permissions to give them and you have an online workspace where you can share your work and draw on the contributions of other members in the Cellucidate community.  To learn more about sharing cBooks, go to our tutorial page and watch the third video. There is also a bookshelf dedicated for cBooks shared with you so you can quickly navigate to them and keep them organized.  And to effectively communicate with other users when working in Cellucidate, you now have the ability to send and receive messages to them through the system. 

Status Bar
When viewing a cBook there is a new status bar located at the bottom of the page that enables you to better manage your work.  With the status bar you can:

• See which collaborators are simultaneously viewing the cBook
• View the models and rules of a cBook by navigating to specific pages using the contact map
• See jobs running in the cBook
• View updates to your cBook
• Navigate to messages you have sent to and received from other users

Other recent improvements to Cellucidate include the ability to import Kappa files and download a snapshot of the information generated from a simulation in a CSV file.

We are always looking for ways to make the Cellucidate experience better for you.  Email your comments to feedback@cellucidate.com or add them to the Feature Requests topic under Support in the Discussions forum in Cellucidate.

Transparent Models

By Isha Antani and Ty Thomson

Understanding and interpreting a differential equation model published in a paper can require one to go beyond merely copying and running the model in a simulator. Sometimes, running someone else’s previously published model generates data that do not agree with the data published in the paper. A researcher who encounters this issue has to go back to the original model and pick apart each and every term in the differential equations to make sure that the term was correctly copied or translated.

Because the original model is often just a text description, and not an executable file, it can be difficult for the researcher to assess whether differences between their simulation results and the published results are due to an error of their own, or due to an error in the original paper.
Unfortunately, errors in published models are not as uncommon as one might hope, perhaps because of the onerous task of re-implementing models during the peer review process.

Even when the original model is available in an electronic and executable format, it can be difficult to read through the model (which is often in the form of computer code) to discern why the two sets of results do not agree.  Perhaps there is something in the description of the original model that disagrees with the executable model.

For models to be easily replicable, transferable, and understandable, they have to be transparent. In this context, transparency means that the representation of the model has to be easy to read and understand, and should thus contain appropriate documentation.  For example, a viewer might be able to click on an equation in a transparent model to visualize the corresponding reactions and assumptions. Such a transparent system, furthermore, can aid the creator of the original model to easily flag any flaws that they might have introduced while creating the model.

There is an increasing demand for ease of sharing models of signal transduction pathways. As the study of pathways and signaling systems spreads, there are bound to be an increase in the number of papers that have model accompanying them. Therefore, having these models be transparent and easily reproducible is going to become even more of a necessity in the future.