Learning to ignore, rare phenotype transitions, allostery in networks, and processing raw single-cell data.
Irrelevance by inhibition: Learning, computation, and implications for schizophrenia
Individuals with schizophrenia have difficulty ignoring ideas and experiences that most people would treat as unimportant. There is evidence that this may be due to changes in neuronal inhibition, suggesting that inhibitory neurons may be involved in learning to ignore irrelevant inputs. By developing a computational model that learns relevance and irrelevance through changes in the strength of feedforward inhibition, Nathan Insel and colleagues are able to simulate many specific effects of inhibitory neuron dysfunction on behaviour. They also show two computational advantages to this mechanism: (1) if relevance is signalled by the level of excitatory activity, then downstream circuits can easily avoid learning from irrelevant stimuli, (2) relevance learning can occur simultaneously with other types of learning.
Rare-event sampling of epigenetic landscapes and phenotype transitions
Cell phenotypes are controlled by complex interactions between genes, proteins, and other molecules within a cell, along with signals from the cell’s environment. Gene regulatory networks (GRNs) describe these interactions mathematically. In principle, a GRN model can produce a map of possible cell phenotypes and phenotype-transitions, potentially informing experimental strategies for controlling cell phenotypes. Such a map could have a profound impact on many medical fields, ranging from stem cell therapies to wound healing. However, an analytical solution of GRN models is virtually impossible, except for the smallest networks. Instead, time course trajectories of GRN dynamics can be simulated using specialized algorithms. However, these methods suffer from the difficulty of studying rare events, such as the spontaneous transitions between cell phenotypes that can occur in Embryonic Stem Cells or cancer cells. In this paper, Margaret Tse and colleagues present a method to expand current stochastic simulation algorithms for the sampling of rare phenotypes and phenotype-transitions.
Network-level allosteric effects are elucidated by detailing how ligand-binding events modulate utilization of catalytic potentials
Enzymatic rate laws have historically been used to simulate the dynamics of complex metabolic networks with regulated reactions represented by allosteric rate laws. Here, James Yurkovich and colleagues use detailed elementary reaction descriptions of regulatory enzymes that allow for the explicit computation of the fraction of the enzymes that are in a catalytically-active state. The fraction of the enzyme that is in the active state represents the time-dependent utilization of the enzyme’s “catalytic potential,” its capacity to catalyze a reaction. They apply this interpretation to red blood cell glycolysis, examining how three key kinases with allosteric regulation modulate their utilization of their catalytic potential based on ligand-binding events throughout the network in order to maintain a homeostatic state.
scPipe: A flexible R/Bioconductor preprocessing pipeline for single-cell RNA-sequencing data
Biotechnologies that allow researchers to measure gene activity in individual cells are growing in popularity. This has resulted in an avalanche of custom analysis methods designed to deal with the complex data that arise from this technology. Although hundreds of analysis methods are available, relatively few deals with raw data processing in a holistic way. Luyi Tian and colleague’s scPipe software has been developed to fill this gap. scPipe is the first fully integrated R package that deals with the raw sequencing reads from single cell gene expression studies, processing them to the point where biologically interesting downstream analyses can take place.