A framework for evolutionary systems biology

BMC Systems Biology

Table 3 Approaches towards understanding biology.

latin	meaning	strength	weakness	analogy¹
in ratio	analytic model	well understood, precise predictions or approximations; can falsify intuitions and hint at simulation errors; can explain data if mechanistic	limited to simple models by mathematical tractability	hard, dry bone
in silico	simulations of more realistic models	can be very realistic; can use more observations than analytic models to make better predictions; can falsify approximations and intuitions; can explain data if mechanistic	sometimes too hard to understand; computing can be costly; some heuristic models can predict data without explaining	flesh
in vitro	experiment without anything alive	precise molecular observation and manipulation possibilities; can falsify models	can be expensive; extrapolation to in vivo is not always possible; complexity limits	food to eat
in vivo	laboratory experiment with living cells	controlled environment allows specific manipulations; can falsify models	relevance for natural settings not always clear; limited mechanistic understanding	water to drink
in natura	observation of organisms in their natural setting	get information on actual natural processes; can falsify models	either only historic or usually limited by ~3 year funding periods; limited mechanistic understanding	air to breathe
in tuitio	ask good questions	very cheap and fast; all ideas start here	is no scientific proof in itself	spirit with good ideas

¹ As analogy, think of a biologist made of flesh and blood. Just bones and flesh are dead unless they breathe air, drink water and eat food. This illustrates that good theoretical models need to be designed to incorporate experimental data in order to 'come alive'. A good intuition is needed to develop such models. Heuristic models can be very good in predicting observations, but true understanding grows only when models reflect true causalities.

ISSN: 1752-0509