"He argues that the study's authors examined only microbial composition, not function — and microbes can change function rapidly, making it hard to predict how

"He argues that the study's authors examined only microbial composition, not function — and microbes can change function rapidly, making it hard to predict how they perform solely on the basis of the genera of bacteria present. A 2011 study found evidence that Clostridium bacteria in panda guts contained genes that resembled those known to produce enzymes that break cellulose down into simpler sugars." It sounds the authors identified the bacteria flora in the panda gut via feces, but not the bacterias' specific gene expressions which may vary even if they are known suited more for carnivore digestion (e.g., but may be specialized to digest plants cellulose in panda gut environment). For those involved in genomics, I'm curious as to how one can obtain the RNA expression (RNASeq) of multiple bacteria present in panda gut; and how do they go from the raw sequence data sequenced from the multiple bacteria composition to match the specific gene and then also its host bacteria?