Two theories that have been making the biological rounds have been group selection which argued for evolution and selection operating "for the good of the group" and the "selfish gene" which argued that everything (including cooperation) was ultimately based on selfish actions.  The latter being especially pronounced in giving rise to all manner of explanations that argued for genetic representation in the future as being the ultimate criteria to which an organism would act.

We now have an article that uses a similar explanation in describing "exploding bacteria".

This seems like an awfully egalitarian act, especially for a bacterium, but the paradoxical reason behind the suicide is to increase the deceased's chances of  leaving descendants. That ought to be pretty hard when you've just blasted yourself to bits, but according to ETH team-leader Fredrik Inglis, the behavior is likeliest to occur in “clonal” bacterial communities, in which all individuals share the same genes. In this situation, it doesn't much matter who survives to divide and who doesn't, since the whole reason all creatures — ourselves included — are impelled to reproduce in the first place is to pass on their genes.
http://ecocentric.blogs.time.com/2011/07/29/exploding-bacteria-self-fertilizing-bugs-and-other-cool-critters/

Of course, it becomes easy to see that this offers no coherent explanation with respect to the "selfish" gene any more than arguing that this act of suicide is being committed for the "good of the species".  In fact every instance, ranging from kin selection examples to altruism, invariably results in either interpretation providing an equally reasonable explanation for observation.

We already know that group selection is true to some degree since there must exist a minimally reproducing pair in any sexual organism, so (by definition) selection must ensure survival of the group.  To argue that the group doesn't influence the genome is to miss the picture regarding the role of sexual selection in such species.  Even in single-celled organisms, there isn't much to discuss until a minimum population density exists.

The selfishness argument is a bit more difficult since there are few examples that stand up to scrutiny regarding its existence or viability.  Since communication even occurs at the single-celled organism level, it is hard to argue that such an evolutionary step can be taken without cooperation being a fundamental aspect of biological development.  While it should be clear that self-interested behavior can occur, so that cooperation doesn't equate to self-sacrifice, the concept of selfishness has virtually no examples that result in a viable organism.  In those cases where selfishness is demonstrated (i.e. segregator-distorter genes), the organism invariably dies.

Actually, if there wasn't so much effort being expended on rationalizing these goofy positions, it just might occur to someone that this is a potential example illustrating how a colony can break down in a manner similar to what we see with cancerous tumors in multi-celled organisms.  After all, cancers occur when normal cells no longer cooperate with those around them and while this example may appear unusual, perhaps the correct interpretation has little to do with genetic reproduction but rather illustrates a fundamental breakdown that simply leads to arbitrary destruction.