Skip to content. | Skip to navigation

Personal tools

Sections
You are here: Home / About / Why does Evolvix exist?

Why does Evolvix exist?

Key forces driving development of Evolvix in a nutshell: math matters and abstractions are an art.

Life is complicated. Life is dynamic. Life exceeds our limitations. Biologists struggle with this complexity of life, whenever they want to understand more about the mechanisms that drive the dynamics of life. If this only was as simple as rocket science! Modeling helps us to understand this complexity and Evolvix is being designed to make modeling easy. That's it in a nutshell. Read on if you want to know more.

 

Why models? 

Biologists have spent decades and centuries describing details of diversity at the molecular and organismic level. This is a necessary first step, but describing is not understanding:

  • to truly understand a biological system, we need to understand the key mechanisms that drive it;
  • to identify the key mechanisms we need good overviews of everything that is going on.  

Such overviews are mental maps that bring together many different observations and specify cause and effect. These mental maps are often called 'mechanistic models' or short 'models'.  Knowing the "why" does not cease to fascinate. Such models provide an explanation of why we see what we see at the surface1.

 

What are good models? 

The quality of models varies from

  • very vague verbal speculations to 
  • precisely computable mathematical proofs backed by rigorous observations. 

Verbal speculation can be cheap, but it often contradicts itself and it often is impossible to test by observations. Some of such contradictions are easier to spot than others and performing tests of models using observations of reality can be challenging. Good models are reliable:

  • reliable models do not contradict themselves;
  • reliable models are well grounded in observations of reality;
  • good models are reliable and also explain something that is not already completely obvious anyway.
How then can we construct good models? Do we need to start from scratch every single time? Not necessarily, if we can use a model description language.

 

Why model description languages? 

Using a particular language makes it easy to say some things and hard to say others. Using a shared dictionary of defined words eliminates the need to evaluate different ways of expressing certain concepts. If that language shares a dictionary that both humans and computers can understand, then humans can describe models which can then be analyzed by computers. Research in computer science and mathematics has shown that

  • appropriately defined formal model description languages can be great tools for constructing models that are free from whole classes of contradictions; 
  • computers can then be used for detecting likely modeling errors and the computation of non-obvious predictions using state-of-the-art simulation methods. 

These methods are truly impressive, but often require models to be constructed in ways that most humans experience as difficult or tedious (depending on skills). This makes it hard to write or understand such models and unnecessarily limits the use of modeling in biology. 

 

Why Evolvix? 

Just because. Here we could bore you to death with

  • a long-winded comparison of the many existing modeling tools, programming languages, simulation libraries and data processing utilities or
  • many detailed descriptions of the features we would be missing if we were to use only the best of these existing tools or
  • arguments about how we think a modeling environment could be made better by doing xyz and why we aim to do that.

  

However, at the end of this long discussion it would all boil down to this:  

Modeling is an art.

   

Each piece of art has its style. Likewise, each model has a modeling style determined by its modeling environment that was designed to support this style. Thus:

Designing a modeling environment and an artistic style is, well, ..., essentialy an art.

   

Why art? Basically, because reality would blow our minds if we would have to take it all in at the same time. Cultural art and scientific models give us the ability to focus; the art is in choosing what to focus on. The similarities are striking and have profound implications. Both, artistic styles and modeling styles 
  • provide a very particular view on reality2 that makes some aspects more explicit than others;
  • have to capture some aspect of reality particularly well and interpret it in interesting ways to be successful;
  • have to be either beautiful or realistic to some degree to find an audience;
  • are hard to combine with other styles without (usually) creating a disorganized mishmash or (rarely) a distinct new style;
  • will always be created by those feeling the need to describe new aspects of reality in new ways. 

So why does Evolvix exist? In a nutshell, its mission statement says it all:

 

Evolvix exists to make rigorous modeling easy, starting with biology.

 

You can read more about what specifically we mean by this on the Evolvix mission statement page.

 

 

Footnotes:

1 If we keep digging deeper into the "why" of observations, we often take one or more mechanistic models for granted, declare they explain the "how", and then redefine the "surface" to now include these models before asking "why" key features of these models are the way they are. While there is no bottom to such digging, it is clear that models play a key role in shaping our understanding of the world. These models better be reliable! Here we focus on 'mechanistic' models, which can sometimes be contrasted to 'predictive' models that focus on the ability to predict results from a given range of starting points that has been previously seen in the form of 'training data'. Such predictive models may not help understand phenomena they predict.

2 In this list, "reality" can obviously refer to physical or abstract realities.