McGill University researchers report on a machine learning method called Bound2Learn, designed for detecting DNA-bound proteins. The team notes that the "time [DNA-bound proteins] spend bound to DNA provides useful information on their stability within protein complexes and insights into the understanding of biological processes." With that in mind, the authors came up with the machine-learning and single-particle fluorescence tracking-based Bound2Learn approach to assess binding activity and dynamics for DNA-bound proteins — an approach they demonstrated in Escherichia coli and Saccharomyces cerevisiae model organisms. More broadly, they add, the Bound2Learn method "will be easily applicable to other experimental models, including mammalian cells."
Meanwhile, investigators from the Southern Medical University, Jinan University, and other centers in China share a CRISPR gene editing-based method for activating specific gene targets in vivo. The "compact and robust miniCas9 activator" (miniCAFE) approach hinges on an engineered form of Cas9 combined with targeted transcriptional activators, the team explains, and was successfully used to activate specific genes in Caenorhabditis elegans worm models, mouse liver, or human cells. Based on their results so far, the authors argue that miniCAFE "holds great therapeutic potential against human diseases."
Finally, a team from France outlines an "optimized pipeline of CLIP-seq data analysis," or optiCLIP, system for teasing out reproducible microRNA binding sites with the help of cross-linking immunoprecipitation and high-throughput sequencing (CLIP-seq) and game theory strategies. When they applied optiCLIP to CLIP-seq datasets focused on the Ago2 Argonaute protein in mouse and human samples, for example, the investigators tracked down almost 100,000 miRNA binding sites, subsequently using a quantitative miRgame model to tally the extent of miRNA occupancy at the Ago2 peaks considered. "[W]ith increasing number of high-throughput data for miRNA targeting, our method is providing a powerful tool to interpret the dynamics and the complexity of miRNA-dependent post-transcriptional gene expression," they write, adding that "it should be possible to extend our model to integrate post-transcriptional regulation by other regulators."