With increased research funding for Alzheimer’s disease (AD) and related disorders across the globe, large amounts of data are being generated. Several studies employed machine learning methods to understand the ever-growing omics data to enhance early diagnosis, map complex disease networks, or uncover potential drug targets. We describe results based on a Target Central Resource Database protein knowledge graph and evidence paths transformed into vectors by metapath matching. We extracted features between specific genes and diseases, then trained and optimized our model using XGBoost, termed MPxgb(AD). To determine our MPxgb(AD) prediction performance, we examined the top twenty predicted genes through an experimental screening pipeline. Our analysis identified potential AD risk genes: FRRS1, CTRAM, SCGB3A1, FAM92B/CIBAR2, and TMEFF2. FRRS1 and FAM92B are considered dark genes, while CTRAM, SCGB3A1, and TMEFF2 are connected to TREM2-TYROBP, IL-1β-TNFα, and MTOR-APP AD-risk nodes, suggesting relevance to the pathogenesis of AD.

Neurons have evolved a number of unique protein-coding genes that regulate trafficking of protein complexes within small organelles throughout dendrites and axons. Neuron-specific gene 2 (NSG2) encodes for one of the most abundant proteins in the nervous system during perinatal development. NSG2 belongs to a family of small neuronal endosomal proteins but its function has remained uncharacterized to date. Here, we show that NSG2 is found in discrete punctae restricted to the somatodendritic arbors of developing mouse and human neurons, and a significant proportion of NSG2 punctae colocalize with postsynaptic HOMER1 and surface-expressed AMPA-type glutamate receptors (AMPARs) at excitatory synapses. Immunoprecipitation revealed that NSG2 physically interacts with both the GluA1 and GluA2 AMPAR subunits in mouse brain. Knock-out of NSG2 in mouse hippocampal neurons selectively impaired the frequency of miniature EPSCs (mEPSCs) and caused alterations in PSD95 expression at postsynaptic densities (PSDs). In contrast, NSG2 overexpression caused a significant increase in the amplitude of mEPSCs as well as GluA2 surface expression. Thus, NSG2 functions as an AMPAR-binding protein that is required for normal synapse formation and/or maintenance, and has unique functions compared with other NSG family members.

Stem cell‐derived neurons from various source materials present unique model systems to examine the fundamental properties of central nervous system (CNS) development as well as the molecular underpinnings of disease phenotypes. In order to more accurately assess potential therapies for neurological disorders, multiple strategies have been employed in recent years to produce neuronal populations that accurately represent in vivo regional and transmitter phenotypes. These include new technologies such as direct conversion of somatic cell types into neurons and glia which may accelerate maturation and retain genetic hallmarks of aging. In addition, novel forms of genetic manipulations have brought human stem cells nearly on par with those of rodent with respect to gene targeting. For neurons of the CNS, the ultimate phenotypic characterization lies with their ability to recapitulate functional properties such as passive and active membrane characteristics, synaptic activity, and plasticity. These features critically depend on the coordinated expression and localization of hundreds of ion channels and receptors, as well as scaffolding and signaling molecules. In this review I will highlight the current state of knowledge regarding functional properties of human stem cell‐derived neurons, with a primary focus on pluripotent stem cells. While significant advances have been made, critical hurdles must be overcome in order for this technology to support progression toward clinical applications.