Table 2

Points on different levels of the adaptive landscape



in the



A molecular structure

in the

space of genotypes


A molecular function

in the

space of molecular structures


A computable emergent property2

in the

space of molecular functions


A computable fitness correlate

in the

space of computable emergent properties


An observable fitness correlate3

in the

space of computable fitness correlates


The fitness of an individual

in the

space of observable fitness correlates


The mean fitness of a population

in the

space of the fitness values of all individuals in the population4

1 The height and plane of a point on each level can have one or many dimensions.

2 This level can have an arbitrary number of sublevels reflecting the hierarchical nature of biological systems. For example 3.1 could compute cell properties based on molecular functions, 3.2 tissue properties based on cell properties, 3.3 organ properties and so on, as needed.

3 This level is a simple 1:1 mapping, if the LFCH can be accepted. In all other cases it may be used to provide some heuristic quantitative link between observed and predicted fitness correlates.

4 The plane has as many dimensions as there are individuals in the population. Providing additional information on genotypes or phenotypes allows a massive reduction of dimensions, producing the traditional views of the adaptive landscape that show the mean of a population in genotype frequency space or phenotypic trait space.

Loewe BMC Systems Biology 2009 3:27   doi:10.1186/1752-0509-3-27

Open Data