Hi all, I have gene micro array data and I want to draw network out of this data. Now I am confuse about two techniques either Markov random field network or correlation based network. Please guide me what are pros and cons of both methods.
These are two different things. One is a graph and the other is a probabilistic model. Correlation-based network refers to a graph whose adjacency matrix is a correlation matrix, i.e. the edge weights are correlations between the nodes. A Markov random field is a type of graphical model, i.e. a graph that captures the relations between random variables. Markov random fields are the undirected equivalent to Bayesian networks. In a Markov random field, the nodes are random variables and the edges represent the interactions between these variables.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
If I want to make a biological network then which graph is more appropriate?
It depends on what your data is and what question you want to address with it.
Can you please explain that for which data Markov random field is suitable and for which data correlation based network is more appropriate? I am working on gene expression data.
It would be more efficient if you stated what the question is that you're trying to address with this data.
I have to make a disease associated network using gene expression data. I want to make a gene co-expression network using correlation, rather than Markov network but could not find reason that how correlation based network is better than Markov network. I am a bit confused about it. Or why should I continue with Markov network if correlation is not appropriate.