Systems Biology

Systems biology is the study of the interaction biological components and how they function or behave together. The system may be modelled mathematically and computationally.

It takes a holistic approach vs the reductionist approach to understanding. An example of a reductionist approach is genomic biology, which identifies all genes and attempts to build up the individual proteins to describe the system. It studies the system at the molecular level. Good for understand parts in detail.

It is now possible to take an holistic approach because:

  • better understanding of biological mechanisms (causally interacting parts)
  • complete genomic data makes it possible to know all parts
  • omics data allows us to profile or view different snapshots of the system.
  • computation analysis aided by ontology and other standards.

Biological systems can be viewed at different levels

  1. organism
  2. organ
  3. tissue
  4. cell
  5. subcell
  6. pathway
  7. protein complex
  8. gene

Each of these levels can be related. e.g. proteins transcribe genes, proteins are translated from genes.

A entity can be studied at multiple levels.

Systems Biology Loop

Build up your model

  1. hypothesise
  2. experiment
  3. gather new data
  4. construct model
  5. analyse model
  6. gain insight from model
  7. repeat and refine


Study of the metabolome, or small molecules. The metabolome is a snapshot of the small molecule products in a cell.

Cells will contain specific small molecules or chemical fingerprints due to their cell processes.

The helps determine the system within a single cell.

Metabolomics Advantages

It is dependent on proteome (and previous transcriptome and genome). Small changes in proteome may lead to more easily recognised changes in metabolome.

Measured faster, cheaper and easier than proteins.


Pathways describe an organism's inputs and outputs (what is needs to consume and can produce).

Pathways help determine the effect of a gene or protein deletion.

There may be alternative ways to achieve same outcome. Redundant paths make the organism more robust.

Comparison of pathways between organisms to study evolution.

Drug targets based on knowledge of pathway and interactions.

Pathway construction

Construction is NOT done via studying individual protein interactions. This will be very time consuming.

Metabolic Control Analysis

Models each enzyme with numerical properties (kinetics). Concentrations of metabolites are controlled and measured. A robot controlled by PC can generate and execute several experiments to test out hypotheses.


Heart model

Easy to measure heart beats (electrochemical pulses)

Immune system

Difficult, it consists of multiple levels / aspects

  1. molecules to organs
  2. temporal. changes can seconds to years
  3. spatial. signalling and diffusion
  4. diversity. between molecules, cells, individuals


Standard terminology or vocabulary for information exchange about a topic.

Reduces confusion between various groups.

Help with computer searches.

Gene Ontology

Areas included are species agnostic gene descriptions for

  1. Molecular function (e.g. cytochrome P450 is a monooxygenase)
  2. cellular component (e.g. cytochrome P450 is found in the endoplasmic reticulum)
  3. biological process (e.g. drug metabolism)


BioPAX is an ontology for describing pathways. It includes definitions of classes with properties.

Markup languaged

Structured XML with schemas. MathML, CellML, SBML


open standard

Documents represent mathematical models of cells.

reusable components

can represent signal transduction and pathway

names of genes and metabolites are standardised.

Systems Biology Markup Language (SBML)

SBML is used for computational models of biological processes.

Noble's heart model

what it does model is important to the system and has well characterized behaviour and interaction


Signalling Pathways

Nodes are proteins

Edges are activation or inhibition (e.g. Phosphorylation turns many protein enzymes on and off)

Complex and interwoven.

Gene Regulation network

Nodes are transcription factors

Edges are regulatory interaction. It is either activation or repression of a gene. Transcription factors may regulate themselves (autoregulation).

Often mRNA cannot measure gene expression of transcription factors as they are very low.

Metabolic pathway network

Nodes are metabolites

Edges are conversion of metabolites by an enzyme.

e.g. The Leloir pathway describes the catabolism of D-galactose.

Protein - Protein interaction networks

Nodes are proteins

Edges are physical interaction between proteins.

Quantitative modelling

An edge tells you about a relation but may not give enough detailed information. e.g. KEGG metabolic pathway will show you relationship, but not quantitatively.

Experimental data will help.

  • spring/hook/wire example with part counts helps determine the components
  • understand of how spring/hook/wire components interact.

Law of mass action

Mathematical model for prediction behaviour of solution in equilibrium. Requires both

  • equilibrium aspect, which is the concentration at equilibrium
  • kinetics aspect, rate equation or how fast it will change to reach equilibrium.

In chemical kinetics, the rate is proportional to the product of the reactants:

$$ A + B = C $$


$$ \frac{d[C]}{dt} = k[A][B] $$

where [x] is the concentration of x

and k is a constant.

Consider the reversible reaction

$$ S_1 + S_2 = 2P $$

The forward reaction rate is v+, backward reaction rate is v-

The overall forward reaction rate is v = v+ - v-

It can be rewritten as

$$ S_1 + S_2 = P + P $$

The concentration of S1 will change at rate of

$$ \frac{d[S_1]}{dt} = - v_+ + v_{-} $$

Substituting v_ and v+ with the law of mass action

$$ \frac{d[S_1]}{dt} = -k_+[S_1][S_2] +k_-[P][P] $$

where k+ and k- are constants.

ref: MassAction.pdf from

Michaelis-Menten kinetics

$$ E + S \, \overset{k_f}{\underset{k_r} \rightleftharpoons} \, ES \, \overset{k_\mathrm{cat}} {\longrightarrow} \, E + P $$

where E is an enzyme

and S is the substrate

ES is the enzyme substrate complex

P is the product

Rate are given by

$$ v = \frac{d [P]}{d t} = v_\max \frac{[S]}{K_M + [S]} $$

where vmax is when all enzyme is bound to substrate.

and KM is [S] when the rate is half of the maximum rate.

Lactococcus lactis system

Gluclose -> G6P -> FBP -> 2 PGA -> 2 PEP -> 2 pyruvate -> 2 lactate

PGA and PEP are kept around

The intricate side of systems biology.