A common creationist hypothesis is that pathogens arise by degeneration. Part of that probably comes from a quasi-theological notion that all changes in creation are bad. Everything in the beginning was perfect, and it's only gotten worse since then. It's an old idea found in the early works of George McCready Price, for example, but its practical relevance is largely untested. How to test it isn't even clear: What does "degeneration" look like? Answering that question necessitates having some specific knowledge about the original state, which is kind of impossible to know, right? What does a perfect pathogen (namely one that is not a pathogen) look like? I don't know. It's hard to say.
A few years back, I tried testing a simplistic model of pathogen origin. I hypothesized that "degeneration" might manifest as less DNA. In other words, if a bacterium lost some of its DNA that codes for important metabolic process (say, amino acid synthesis), it might become a pathogen in order to cope with that loss (by mooching off other creatures). How do I know what the original state was? I tried comparing pathogenic bacteria to close relatives that were not pathogenic. I found that around half of the pathogenic bacteria had more DNA than their close relatives. This comes as no surprise to microbiologists, but it was something I wanted to document for creationist biology: Not all pathogens are characterized by small, obviously degenerated genomes. This was published as an abstract in the Occasional Papers of the BSG.
A recent paper published in Genome Biology expands on this point. The authors Stabler et al. looked at genomes of Clostridium difficile. Infection with C. difficile can give you mild to extremely severe diarrhea. A new variety of C. difficile strain 027 has emerged in the last five years, producing more severe symptoms and a higher mortality rate. This strain 027 has not been associated with epidemics in the past, but this new variety is hypervirulent. Stabler et al. sequenced the genomes of non-epidemic and hypervirulent 027, and compared both sequences to the genome of C. difficile 012. The hypervirulent strain had 47 genes not present in either of the other difficile genomes. The majority of those genes were of unknown function, but some of them were associated with drug sensitivity, implying a possible role in disease. The point here is that once again, we have the emergence of a pathogen by acquisition of DNA.
Now I suppose some might call the acquisition of genes a "degeneration," but that seems to be motivated more by a desire to preserve the "all change is degenerative" mantra than by any evidential reason.
Stabler et al. 2009. Comparative genome and phenotypic analysis of Clostridium difficile 027 strains provides insight into the evolution of a hypervirulent bacterium. Genome Biology 10:R102.