You are here: Home / About / Welcome to the Modeling of Evolution!

Welcome to the Modeling of Evolution!

Understanding change over time is at the heart of modeling evolution. Delegating trust is at the heart of modeling.

Have you ever wondered why life is constantly changing?

How the different parts fit together?

How it all works?

Welcome to biology in the 21st century. One of the big goals of biology is to understand the mechanisms that allow life to do what it does and to quantify it all with increasing precision. As biologists we integrate this understanding into we call 'models'. These models can take many different forms from verbal stories and causal diagrams to mathematical equations and computer programs. In a nutshell, we can think of them as maps that can help us navigate the complexities of life - that is if we have the right map for the question we are asking. In case you wonder what that means, try to use a geological map of ocean floors for finding the next shopping mall and you will get the idea.

This means it is just as important to use the right model to answer a question, as it is to use the right tool for a given task. Not following this principle can easily hurt us, others, life in general and the environment necessary for life to prosper.


Can you drive a car?

Let's stick with that analogy for a moment. Driving a car is not overly complicated. Most of us can easily learn this. However, cars are quite dangerous instruments and can cause a lot of harm if used in the wrong way. This does not stop us from using cars. On the contrary, it motivates us as a society to make sure that anybody who drives a car understands and follows a few important rules while out on the streets. We even have tests for ensuring that newcomers are up to the game and we give them a "drivers license" if they pass.


Can you build a car?

As easy and fun as it can be to drive a car, building a car is an entirely different matter. Do you know how to build a modern car? Do you know anyone who can do it?

I don't.

Do we have to know how to build a car in order to be good drivers? Of course not! There are a few commonalities known by all who build cars and all who drive cars: Where the breaks are and what they do, how a steering wheel works and so on. These agreements are important, as they provide drivers of cars with a common language telling makers of cars what the drivers expect their car to do next. In computer science such agreements are also called "user interfaces". However, as important as they are, they do not help building an engine or master the gigantic amount of arcane information and skills necessary for building a car.

In other words, we have become accustomed to ride fire-breathing dragons on 4 wheels at great speed, even though we don't really understand what is inside and we all know reliable reports of how they can go up in flames or smash people like flies to a wall! Why are we not worried?


Why are you allowed to drive a car, if you cannot build a car?

Basically, because over decades we have built a chain of trust with many checks and balances that is designed to ensure that the many experts we pay to build cars each do their job well. We demand evidence that the experts are indeed experts as well as independent certificates that the cars pass very specific tests before we allow them to be used in public. Most societies also encourage the existence of many different car manufacturers to encourage them to build the best possible cars (or be put out of business by competitors with a better track record), in addition to a rather free flow of information on problems with specific cars.


Driving models and building models in science ...

... is not different. It is often rather easy to use the results of scientific models that were rather difficult to build or test. Just think of the science in cars or computers! Building or testing such models often requires the use of a very diverse set of other models and skills that can only be mastered by extraordinary long training. This training can sometimes require a lifetime! As a society we have found it useful to divide labor such that some people dedicate their life's work to this extraordinarily complicated task of building and testing new scientific models that help us to better understand particular aspects of life, the universe or everything. We call them 'researchers' or 'scientists' and we expect them to tell us the truth to the best of their abilities. To make sure we don't get fooled, we have built a chain of trust with many checks and balances. In math this chain has been built over thousands of years. In biology and physics over hundreds of years. In some new areas like computational science we are right in the middle of working to build this chain of trust.


How trust works

Trust can't be demanded, it can only be earned. Earning trust is about demonstrating reliability repeatedly and providing good reasons for why the same reliability will remain available. That is why reproducibility is so important for science. Here is how this plays out in science: As with cars, there are many different manufacturers of scientific models, which we usually call 'scientists' or 'schools of thought'. There is a more or less free competition in the heads of scientists between ideas about how to explain some aspects of the world. We often call these ideas 'models' once they are fleshed out a bit more. If trust indeed can only be earned, then new ideas have to earn their trust by demonstrating their reliability repeatedly. It is very important to recognize and define precisely the boundaries of these ideas, as no model applies to everything. Testing models is an important part of the work of a scientist. It means identifying the weaknesses of a model with the goal of weeding out models that are not useful for understanding our world. After all we want to keep the most important models and spare ourselves all others. Many scientists and an open publication process based on reviews by many other scientists who all would like to explain the world better, translates into a rather short shelf-life for unreliable models. While some errors take longer to uncover than others, this random poking, searching and testing in lots of different places has proven to be amazingly effective in getting us closer to a more reliable understanding of our world. However, the world is complex and especially in biology we have a long way to go, despite the tremendous progress we have made.


The big picture ...

Occasionally decades or centuries of research produce a crystal clear conceptual model that describes how many aspects of our world that would otherwise be rather confusing. Newton's mechanics and Einstein's relativity are such models. Chemistry has the Periodic Table of the Elements and Biology found Cells, Genes, DNA and many other conceptualizations extremely useful. However, for biology one conceptual model stands out in its complexity and universality: Evolution. If you think it is complex to understand how to build a car, try evolution. It makes building car's look like child's play.

That is why this website is dedicated to furthering a mechanistic understanding of evolution and supporting work that enables that.


... and its many facets:

To enable a mechanistic understanding of evolution requires mechanistic and quantitatively precise models in:

  • population genetics 
  • ecology and environmental factors
  • life history and physiological trade-offs
  • cell and developmental biology
  • molecular systems biology, biochemistry and structural biology
  • the underlying chemistry and physics
  • simulation technology, process algebras, distributed computing, databases, programming languages and computer science
  • mathematics, probability theory and statistics
  • and many more details all across the spectrum of scientific disciplines.

This makes the quantitative study of evolution extraordinarily challenging and exciting!


If models are like maps and cars are like simulators, then responsible driving is up to us.

We hope this website contributes to more sensible 'model driving' in this world.
We have only one planet to share among us, our children and all animals.
We shouldn't waste it by trusting unreliable models.

Death to bad models, not to living species.