Our aim was to explore causal networks between milk fatty acids by applying for the first time the adapted IC algorithm as presented by Valente et al. [ 12] to 14 highly correlated traits.
It is important to note the difference between the knowledge-based causal network models used in this work and the networks constructed from expression data using network inference approaches [ 30].
Interference between dynamically similar regulators in causal network inference is a general problem.
This is similar to causal network analysis, which is an approach for identifying relationships between concepts in a graphic and time-ordered fashion (Miles and Huberman [1994]).
The genes within the causal network represent nodes and the edge that defines the biological relationship between two nodes is represented as an arrow signifying regulation.
"Severing one link in a causal network," they write, "still leaves the rest of the network intact".
The causal network does not specify how each causal factor produces its effects, though.
Causal network determination relies on causal signals leaving a signature in the gene expression dynamics, essentially a correlation between X t + 1 and X t.
If the causal network is known, Pearl's do-calculus allows estimating the strength of the causal effects between pairs of random variables from observational data (Goldszmidt and Pearl, 1992; Pearl, 1995).
The resulting partially causal network is shown in Figure 3.
