Probabilistic inference in anMRF involves computingMarginal distributions over the randomvariables in the graph. Graph formalismis an effective and efficient representation for multivariate independence structure for both model construction and for inference. The use of graphs to represent independence structure in multivariate probability models has been pursued in a relatively independent fashion across a wide variety of research disciplines. Traditional graphical models decompose the joint distribution as a product of functions of subsets of variables. However, a number of rigorous approximation algorithms have been devised for performing inference in MRFs. Factor graphs are more useful for describing models that involve a large number of overlapping relationships between variables.When compared to Bayesian Networks (BNs) andMarkov RandomFields (MRFs) Factor graphmodel decomposes interactions between variable. While functional relationships between variables in BNs and MRFsmust be determined by identifying parent-child clusters or maximal cliques, Factor graphs explicitly identify functional relationships. Any Bayesian network or Markov random field can be represented as a factor graph. Belief propagation algorithmis used for finding theMarginal probability of the any hidden Variable conditioned on the observed variable. The algorithm is designed by passing real valued functions called messages along the edges between the nodes. This paper analyses the application of the belief propagation onBiologicalMarkov randomfieldwith an example.