One example would be the Konijnendijk et al. (2011) study of Lake Victoria cichlids in International Journal of Evolutionary Biology (http://dx.doi.org/10.4061/2011/616320).  An early community genetics analysis, published by Wares (2002) in Molecular Ecology, also included teleosts (DOI: 10.1046/j.1365-294X.2002.01510.x)

Thanks for your answer Marc! The study of Konijnendijk adopts a particular meaning of community genetics, that is, the joint analysis of genetic variation in multiple species, and gene flow and differentiation.

I would like to see is studies that employ information on genetic variation and genetic variants in a species to understand how the ecological interactions among species are shaped. For example, how are predation interactions affected by the predator (or the prey) genotype ? I remeber some "community genetics" studies of this type on plant-consumer interactions for example. But I have never heard of any one on fish....  

If I'm reading your question correctly, I think you need to look to genetic studies on teh midas complex of cichlids in Central America. There was a paper looking at how different prey choice and predator behaviours led to the expression of three distinct morphs of what was originally one species. I think they used genetics to confirm their observations - speciation in motion. 

Daphnia body shapes alter over successive generations if exposed to predators to make them more robust and "spikier". I think that was on a poster pres. at ESA 2013 so may be in paper form by now.

I would guess that zooanthids, palys or octocorals may be a good area for study of coral genetic variation vs predators as they reproduce quickly, so may be able to express different types when exposed to predation? (or may be clones... I'm throwing an idea off the top of my head here).