On Pharyngula, PZ Myers criticizes the stubborn obfuscations of Michael Behe, who refuses to yield his illogical calculations. Behe says (rightly) that a certain mutation necessary for drug resistance in the malaria parasite has about a 1 in 1020 chance of occurring. But the mutation is also detected in 96% of malaria patients who respond well to the drug; it proliferated widely because, by itself, it had no impact on the parasite’s fitness. The parasite needed another mutation, occurring at a later date, to develop resistance to the drug. Behe rests his case for divine intervention on the basis of bad math; as PZ writes, “It was crude, stupid, and ridiculous when J. Random Creationist was doing it, and it’s even worse when a guy with a Ph.D. in biochemistry, who ought to know better, panders to the mob of creationists who don’t even grasp middle school mathematics by using fallacious operations in probability.” Meanwhile, Orac reports that one of the flu strains targeted by this year’s vaccine “has undergone what is referred to as ‘genetic drift,'” making the vaccine less effective than desired. Yet the vaccine still offers protection against about 57% of circulating strains. On Life Lines, Dr. Dolittle shares research that says consuming caffeine while pregnant can effect genes in the baby’s heart. In total, researchers “identified 124 genes and 849 transcripts that were altered by exposure to caffeine in utero.” And on ERV, Abbie Smith reviews the evolutionary trajectory of HIV, which may be tending toward a ‘truce’ with human hosts.