The Bruggeman lab

Systems Bio Q&A

Systems biology: Question and Answers

Question 1: Why do I write about this?

I am currently teaching a few lectures during the course Introduction to Systems Biology. This course is part of the joint master programme: Systems Biology and Bioinformatics at the VU University and the UvA (University of Amsterdam). Every year, during the start of this course, I realise that the beginning students have a hard time figuring out what systems biology now really is.

The following type of questions pop up frequently;
What is the precise role of mathematical modelling and experimentation in systems biology?
Do systems biologists also carry out experiments themselves or are they theoreticians?
What differentiates systems biologists from molecular biologists, cell biologists and bioinformaticians?
Why is bioinformatics a different “field” than systems biology?
How does systems biology relate to other biological disciplines?

Many students view systems biology as a theoretical discipline, distant from the more familiar biological disciplines such as genetics, biochemistry, molecular biology, and cell biology. You will see after reading this text that systems biology fits well among these disciplines and fills a “gap” in biology.

I have decided to answers such questions on this website. I emphasise that those are my personal opinions and that other systems biologists or life scientists may think very differently about them.

Question 2: What is Systems Biology?

It is not straightforward to give a definition of any discipline — not of genetics, ecology, biochemistry, structural biology, biophysics or soft-matter physics. Generally, disciplines are not independent. They tend to overlap with each other, because they exchange methods and concepts, such that clearcut boundaries between disciplines do not really exist. Moreover, a satisfactory definition of genetics in the 1970’s likely no longer captures the field nowadays. In addition, the relations between disciplines then are no longer valid now. Finally, disciplines may (e)merge and disappear. Science after all is a process.

In my opinion, a characterisation of Systems Biology requires an appreciation of the time when it first arose, around the year 2000. At that time, networks were a hot concept in molecular cell biology and genetics; much more so than it was a decade earlier. Networks were so “hot” because of the introduction of a few new measurement technologies — most noticeably at the time was the microarray technology — and the problems that people were having with interpreting the type of high throughput and large-scale data sets that they generate. Of course, before this time biologists were also thinking about networks but then they were much fewer in number than at the time when systems biology arose. So, biologists were confronted with a new type of problem: “How to make sense of the molecular networks underlying cell behaviour?”.  Scientists familiar with dynamic systems, such as engineers, physicists, computer scientists, and mathematicians, are very comfortable with thinking about how networks function, so they quickly became involved in the data analysis, interpretation and experimental design. This is how systems biology came into existence.

Why is it a different discipline? Let me first note that: I am not particularly fond of having all those borders between life scientists because ultimately we all would like to understand how cells functions in terms of the interacting molecules; because: i. this is the only way forward in medicine and biotechnology, and ii. because cells are amazing systems that I am continuously puzzled by: How is it even possible that they work? How are sensing, self-repair, adaptation, and growth molecularly implemented? Do universal principles exist in biology that characterise those amazing properties of living systems? Understanding how all these awesome cellular behaviours are achieved would greatly help us designing much smarter technology and medicine. Self-repair, self-learning and adaptation are clearly way beyond our current engineering capacity and understanding them better would help many other disciplines as well.

Why is it a different discipline? Systems biology is a different discipline now but I guess it will have merged with molecular cell biology in a decade time or so; when the training of the new generation of molecular cell biologists has adapted itself to the requirements of the modern life sciences, which have to do with being able to study molecular networks.  So, the time is different now because systems biology is still new; it uses/tries-out different methods than normally used in molecular cell biology labs. After a decade or so, those methods that have proven their use, will become mainstream and “assimilated” by molecular cell biology (and systems biology will merge with it, because the end goal is the same). In the same way, as software for protein structure studies, blasting of gen(om)es, and microarray facilities is being routinely used in such molecular labs. However, the informatics element of bioinformatics — the databases and algorithms — likely will not merge with the more molecular experimental life sciences disciplines. In addition, the hard mathematics branches of systems biology — which is still very small  — is also likely to stay more separate from molecular, experimental life sciences.

So, what is systems biology? It is a subdiscipline of molecular cell biology that aims to understand the network-level properties of molecular-interaction networks and sometimes also of cellular-interaction networks. Concepts from physics, engineering, mathematics, and biology are used together to answer network-related problems in biology.

Question 3: Is Systems Biology really more theoretical than other biological disciplines?

No, systems biology asks different questions. Questions about how molecular networks operate and less about how individual molecules (including proteins) function. Biochemistry and molecular biology (including genetics) study the properties and processes catalysed by the (macro)molecules occurring in cell biology. Those molecular disciplines are loaded with theory and concepts as well — and this is often forgotten. They aim at elucidation of general principles about (macro)molecules. They do not focus so much on principes of molecular networks. Networks are the study focus of systems biology, which only started to make sense as soon as we knew enough about macromolecules to make this step up in biological complexity. So, systems biology is logical continuation of the molecular biology agenda to obtain a reductionist understanding of living systems; simply because biology realises now that the answers are only partially related to the molecules themselves and for a large part in their interactions. So, the molecular reaction network of a cell is the next study focus.

My view is that there is nothing fundamentally different in the approach taking by molecular biology and systems biology: it is the search for general principles and their exploitation, but of different things.

What makes molecular biology, genetics, and biochemistry also conceptual and theoretical disciplines? Are they not experimental disciplines? My view is that they use experiments to prove the validity of universal concepts and principles. In exactly the same way as systems biology does. This is illustrated by nearly every finding about macromolecules that has proven to have general applicability in biology and made it into textbooks. Here follows a list:
Gene sequence
The central dogma of molecular biology
The molecular working of transcription, translation, replication, recombination etc.
Operon structure
Fitness, selection pressure, selection coefficient
Transcription regulation
Energy transduction by proteins and transporters (e.g. ATPase)
Kinetic proofreading
Protein domains and folds
Activation of regulatory proteins, by way of covalent modification
Conformation changes
Allosteric interactions
Enzyme kinetics
Signal recognition and transduction
The molecular working of the immune system
Flow of current in axons for neuron-neuron communication
Ion channels
Plasmid systems, bacteriophage life cycles
etc, etc, etc

These concepts are all part of the theory of molecular biology, biochemistry and genetics and what characterises their succes. The experiments were carried out to offer support for these concepts while they were still hypotheses.

So, now that systems biology tries to find similar universalities at the level of networks does not make it a more theoretical or mathematical discipline than existing molecular biology disciplines. Systems biology is in my view the logical next step up in biological complexity: now that we know the molecules, we need to start understanding how they function in networks and give rise to network level functionalities, such as signal transduction, organelle biosynthesis, the cell cycle, meiosis, etc. This is a clearcut continuation of the (reductionist) molecular biology agenda, which ultimately aims at understanding how cells and cellular systems work in molecular mechanistic terms. It does bring with it the use of other “tools”, such as mathematical models.

Question 4: Why do we need so much mathematical modelling in systems biology?

All sciences that carry out quantitative measurements use mathematical tools. When it is about the numbers and not qualitative differences, mathematics pops up for free. Depending on the type of data and research question, this may then involve mathematical models, say based on differential equation, but in other cases this may be equations, statistics, probability theory, etc. For instance, quantifications in genetics led to population genetics that tests the consequences of mutation, selection and drift for the frequencies of genes in populations. Epidemiology has a similar theoretical basis but focuses on the origins and spread of diseases, given numbers of disease occurrences over time and locations. Biochemistry developed enzyme kinetics, which includes a classification of different enzyme mechanisms, their discrimination on the basis of experimental data, and the estimation of their parameters given experimental data. So, many of the concepts mentioned above have a precise quantitative definition to allow for their measurement across different systems and species (“unification”): cooperativity, allostery, ion channel kinetics, selection coefficient, energy transduction, transcription initiation, etc. Models and equations appear naturally in all quantitative sciences, regardless of whether you are thinking about life sciences or not.

Mathematical models, based on differential equations or stochastic processes, appear in systems biology because those formalisms are very appropriate when the molecular networks of living cells are studied. The influences that individual molecules exert on each other, to cause them to change state and activity, and eventually cause the changes in concentrations of particular molecules, is a basic feature of such networks that we would like to understand. We require mathematical models to achieve this understanding. An inherent property of networks is their dynamics and this can only be understood with mathematical models. Of course, you can “just” measure the dynamics and report it in a paper — as is often done — but this does not give you any understanding of the molecular interactions that gave rise to this behaviour, only the latter activity leads to understanding of the network whereas the former is “only” monitoring. Note that we need this additional molecular understanding because in the end all interventions in biotechnology and medicine are molecular!

So, mathematical models are used in systems biology because the aim of this discipline is to understand how the dynamic molecular networks occurring in cells arises from molecular interactions, which can only be achieved using mathematical models.

Question 5: Why is bridging the gap between genetics, molecular biology and cell biology a challenge?

There exist many different aspects to this question and I am not writing an essay about this, nor do I want to be exhaustive. What we do need to figure out is; what is meant by “bridging the gap”? To understand this we look shortly at the definitions of some well-known and successful life sciences disciplines:

Population genetics: this discipline aims to understand the changes in gene frequencies in populations of organisms in terms of evolutionary processes (selection, drift, and mutations) and the associated molecular mechanisms. The last problem cannot be solved without genetics.
Genetics: this discipline aims to understand the structure of genes, the processes acting on genes, and the consequences of mutations in genes on the proteins it encodes. The last problem cannot be solved without molecular biology.
Molecular biology: this discipline aims to understand how the structure (primary, secondary, tertiary, and quaternary) of macromolecules in cell leads to the activities and properties of those macromolecules that contribute to their functional role in a living cell. The premise of molecular biology is that this knowledge for all molecules is required — so, necessary and not sufficient — for a molecular understanding of cell behaviour. This cannot be achieved without systems biology and bioinformatics.
Bioinformatics: this discipline aims to store, analyse, and relate molecular data and develop the computational (algorithms), software, database, and standards to do this. So, bioinformatics make data computable and relational such that is can be integrated in such a way that we can study large molecular systems active inside cells. Those molecular systems are dynamic and complicated to think about, which is why this cannot be achieved without systems biology.
Systems Biology: this discipline aims to understand how the functional role of molecular networks inside cells emerges from the molecular properties and interactions. The premise of systems biology is that if we repeat this for enough molecular systems that we eventually have a better understand of cellular behaviour in terms of its molecular network. Clearly, this latter endeavour requires cell studies, the realm of cell biology.
Cell biology (including microbiology): this discipline aims to understand the behaviour of cells and how this is molecularly implemented; differentiation, apoptosis, cell division, cellular movement, etc. One important aspect is also the characterisation of the molecular mechanisms and cellular behaviours associated with cellular interactions.
Multicellular biology and ecology: these disciplines aim to understand how the relations between cells and organisms give rise to higher level properties of organs, multicellular organisms and ecosystems.

So, the underlying agenda of the life sciences is to obtain a molecular understanding of life — of organisms and their interactions. Whether this understanding can ever be completely achieved is irrelevant, this is the aim and the  successes achieved so far have had important consequences for medicine and biotechnology; so, this is a successful approach and a fruitful aim. Another conclusion is that the biological disciplines can all be ordered along a hierarchy that bridges the gap from gene to multicellular organism or ecosystem.

Before the introduction of systems biology there existed a “gap” in this hierarchy, it was realised that network-level understanding was missing and required.

Question 6: Systems biology, computational biology, and bioinformatics: what are the differences?

A simple view:

  • Bioinformatics stores, standardises, and inter-relates data to make it computable. So, it deals with databases, software, data standard, ontologies, etc.
  • Computational biology develops the algorithms and software for performing the computations on data.
  • Systems biology uses data and algorithms to identify how networks function and identify network-level principles; for which we may require dedicated computational tools, feeding back on computational biology.

These fields are clearly highly related because they are all in the business of integrating molecular knowledge into understanding of higher-level molecular organisations in living cells. So, their borders are very fluid and individual researchers are rarely confined by them.