In Science this week, Stephen Liggett led work that sequenced and analyzed all known human rhinovirus genomes. Phylogenetic analysis revealed "conserved motifs; clade-specific diversity, including a potential newly identified species (HRV-D); mutations in field isolates; and recombination," his team writes in the abstract. They also found that a hypervariable 5' untranslated region tract may affect virulence.
Two papers got a lot of news coverage this week for linking automation to biology. In one, computational scientists at Cornell University developed an algorithm to search for natural laws of physics in large data sets. They tested their "principle for the identification of nontriviality" by setting the algorithm to get information from moving systems like oscillators and pendula. Without any prior knowledge about physics, kinematics, or geometry, they say, "the algorithm discovered Hamiltonians, Lagrangians, and other laws of geometric and momentum conservation."
In a second paper, Aberystwyth University researchers created a robot, which they named Adam, that could work out the function of genes that play a role in yeast metabolism by observing growing cells. Blogger-scientist Derek Lowe sees a role for robots like Adam in automating drug screening, and the scientists are already working on a similar robot built just for this, called Eve, says a BBC article.
Scientists have also checked into how nuclear receptors convert a global hormone signal into cell-specific gene expression patterns. By coupling let-7 promoters to a luciferase reporter gene, they showed that the nuclear receptor DAF-12 activates transcription of these let-7 microRNAs in C. elegans. These then down-regulate their target gene, hbl-1, to allow developmental progression of epidermal stem cells.