Luminomics (United States)

Abstract Amblyopia is a neurodevelopmental condition of vision affecting 2-4% of children. Patching therapy is marred by low adherence; over half of patients carry residual functional deficits into adulthood. Luminopia, a prescription digital treatment that incorporates therapeutic modifications into children’s programming via a head-mounted display, was shown to improve visual acuity in a randomized clinical trial (RCT) for amblyopia and led to FDA-clearance, available by prescription. This multicenter registry study compared real-world outcomes to RCT results. Among 40 registry patients, similar in age and amblyopia characteristics, amblyopic eye best-corrected visual acuity improved 1.7 lines (95% CI: 1.2-2.2) after 12 weeks, equivalent to RCT gains (1.8 lines; p=0.0375). Real-world adherence (73%) exceeded that for patching. No adverse events were reported. These findings demonstrate Luminopia's effectiveness in clinical practice, offering a binocular alternative to traditional treatments with high adherence and meaningful visual improvements. Trial Registration: NCT06429280

As humanity advances toward long-term space habitation, maintaining crew cohesion, resilience, and utility in confined, high-stakes environments presents a critical challenge. Building on the growing evidence that behavior is deeply rooted in genetic and physiological phenomenon, this study proposes the “Shepherd of Sorting” framework, which is a biologically guided system designed to integrate genetic, behavioral, and physiological data to predict psychosocial compatibility, optimize role assignments, and support reproductive planning among interplanetary crews. The framework is grounded in principles of pleiotropy and gene–environment interaction, employing polygenic risk scores (PRS), multi-omics datasets, and machine learning algorithms to model crew dynamics and inform adaptive health interventions. The proposal proceeds in seven integrated stages: first, surveying the growing literature demonstrating genetic and physiological underpinnings of social, behavioral, and occupational traits; second, outlining the Shepherd of Sorting theoretical framework and its objectives; third, presenting a methodology for system validation in analog space environments; fourth, exploring biodata applications for crew composition, role matching, and mental health differentiation; fifth, introducing a data-driven human breeding and pairing framework designed for population viability in confined, resource-limited settings; sixth, describing the Shepherd platform—developed by the biotechnology company Laminomics—that operationalizes the model through real-time data integration, social modeling, and ethical decision support; and finally, reflecting on broader implications, limitations, and ethical considerations of genetically guided human systems for space and Earth. Preliminary simulations suggest the framework can enhance crew performance by enabling data-driven selection and role specialization, while ensuring offspring fitness and compatibility. By reframing behavioral traits as biologically emergent and context-sensitive, this approach offers transformative potential for space health governance and holds significant implications for psychiatry, workforce development, and public policy on Earth. Future research will focus on empirical validation of predictive models and ethical frameworks for implementation.

Multidimensional data analysis

diffsims is an open-source Python library for diffraction simulations. This minor release provides some new features and internal improvements. The code remains in active development and updating to this version may break existing code, despite our efforts to make it clear when diffsims changes in a "breaking" fashion. See the list of closed PR's and issues avaliable here for details. Overview A range of internal changes have been made that improve speed, accuracy and readability. These most notably effect the generation of rotation list (which now makes use of the package orix) and the organisation of "geometric" (as opposed to "atomic") kinematical spot pattern simulations. Breaking changes Almost all functionality that has been touched in this release may have changed incompatibly with the previous minor version. New Features Kinematical scaling functions can now be specified by the user (#118) Simulation code that previously lived in pyxem has been migrated here (#101) A new class, <code>ReciprocalLatticePoint</code> for storing a crystal structure has been implemented (#111) Improvements Rotation list grids have been substantially improved by making use of orix (#104 and #115) Bugfixes Multiple bugfixes for <code>ProfileSimulation</code> and associated methods (#102 and #119) Misc and Dev The credits now accurately reflect the contribution of @robtovey (#113)