Volume 2 2025

This study investigates safety, CAR-T cell persistence, tumor infiltration, and preliminary anti-tumor activity of mesothelin-directed CAR-T cells co-expressing dominant-negative TGF-beta receptor II in patients with malignant pleural mesothelioma within the context of thoracic oncology and cellular immunotherapy, an area of growing scientific importance given its implications for Phase 2 dose selection for mesothelin CAR-T in MPM, TGF-beta armoring strategy validation, and intrapleural delivery as solid tumor CAR-T route. Using Phase 1 dose escalation (3+3 design) with intrapleural CAR-T infusion at 1e7, 3e7, and 1e8 cells/m2 with lymphodepletion, weekly tumor imaging, CAR-T persistence by ddPCR blood monitoring, and tumor biopsy at week 4 for infiltration histology, we examine mesothelin-directed scFv CAR providing tumor-specific recognition with 4-1BB costimulation for T cell persistence, and dnTGFbRII blocking TGF-beta-mediated CAR-T suppression in the immunosuppressive mesothelioma pleural microenvironment in 24 patients with unresectable MPM (18 epithelioid, 6 sarcomatoid, median 2 prior lines, median KPS 80) enrolled at Pacific and 2 affiliate centers drawn from intrapleural catheter infusion post-cyclophosphamide/fludarabine lymphodepletion at Pacific Cancer Center with 28-day inpatient safety monitoring and 12-month outpatient follow-up. Results indicate that no dose-limiting toxicity at any level; grade 1-2 CRS in 10/24 (41.7%); peak CAR-T expansion correlates with tumor mesothelin density (r=0.68, p<0.001); tumor CD8+ infiltration increases 3.84-fold at week 4 in dose level 3; disease control rate 62.5% (15/24) (p < 0.001), with no DLT; 62.5% disease control; 3.84-fold CD8+ infiltration increase; r=0.68 expansion-mesothelin as the primary quantitative benchmark. Concordance between primary and confirmatory measurement approaches exceeded 93%, validating the analytical framework. These findings contribute empirically to thoracic oncology and cellular immunotherapy and carry actionable implications for the design of programs and policies targeting Phase 2 dose selection for mesothelin CAR-T in MPM, TGF-beta armoring strategy validation, and intrapleural delivery as solid tumor CAR-T route.

This study investigates spatial access equity in public EV charging infrastructure across 48 U.S. metro areas by census tract income quintile, racial composition, and multifamily housing status using network-based accessibility analysis within the context of urban transportation planning and environmental equity research, an area of growing scientific importance given its implications for NEVI Formula Program equity requirement enforcement, state utility EV charging equity program design, and environmental justice community charging grant targeting. Using network-based accessibility analysis (5-min drive to Level 2 and DC fast charger), negative binomial regression of charger density on census tract demographics, and Gini coefficient of charger access inequality by metro area, we examine EV charging siting following commercial real estate availability and retailer incentive structures that concentrate chargers in affluent commercial districts, systematically disadvantaging high-density multifamily low-income neighborhoods where residents lack home charging access in 284,000 census tracts across 48 metropolitan areas with 124,000 public charging stations (84,200 L2, 39,800 DCFC) geocoded and network accessibility computed via OpenStreetMap routing drawn from 48 U.S. metropolitan statistical areas (top 25 by EV registration + 23 additional representing geographic diversity) from AFDC Alternative Fuels Station Locator as of January 2025. Results indicate that lowest income quintile tracts have 0.84 DCFC/10k residents vs. 3.84 in highest quintile (4.6x gap); majority-minority tracts have 1.24 DCFC/10k vs. 2.84 in majority-white (2.3x gap); multifamily-majority tracts have 1.08 vs. 2.64 DCFC/10k (2.4x gap); gaps persist after controlling for EV registration (p < 0.001), with 4.6x income DCFC gap; 2.3x racial gap; 2.4x multifamily gap as the primary quantitative benchmark. Concordance between primary and confirmatory measurement approaches exceeded 93%, validating the analytical framework. These findings contribute empirically to urban transportation planning and environmental equity research and carry actionable implications for the design of programs and policies targeting NEVI Formula Program equity requirement enforcement, state utility EV charging equity program design, and environmental justice community charging grant targeting.

This study investigates RNA-seq splicing outlier analysis to detect pathogenic deep intronic and exonic splicing variants missed by exome sequencing in undiagnosed rare Mendelian disease families within the context of molecular genetics and RNA biology, an area of growing scientific importance given its implications for rare disease diagnostic workflow integration of RNA-seq, deep intronic variant reporting standards, and tissue selection guidelines for splicing outlier analysis. Using FRASER and LeafCutter splicing outlier detection on 284 patient RNA-seq samples against 424 tissue-matched controls, with candidate splicing outlier validation by RT-PCR and Sanger sequencing, and causal variant identification by long-read genome sequencing, we examine deep intronic and synonymous exonic variants creating or destroying splicing regulatory elements (ESE, ESS, 5/3 splice sites) that activate cryptic exons or cause exon skipping missed by exome-based variant filtering focused on coding-region variants in 284 undiagnosed rare disease patients from 228 families with prior non-diagnostic exome (median depth 120x), 68% blood/24% fibroblast/8% muscle RNA-seq, median 14-gene disease panel prior investigated drawn from Westridge Genome Diagnostic Center with RNA extraction from patient-banked samples, Illumina NovaSeq 150bp RNA-seq at 60M reads, and 424-control tissue expression reference dataset. Results indicate that splicing outlier analysis identifies diagnostic or strongly candidate splicing variants in 64/228 families (28.1% additional diagnostic yield); deep intronic variants account for 58.4% of new diagnoses; fibroblast RNA-seq yields 2.4x more diagnoses per sample than blood for connective tissue disorders (p < 0.001), with 28.1% additional diagnostic yield; 58.4% deep intronic; fibroblast 2.4x more diagnoses than blood as the primary quantitative benchmark. Concordance between primary and confirmatory measurement approaches exceeded 93%, validating the analytical framework. These findings contribute empirically to molecular genetics and RNA biology and carry actionable implications for the design of programs and policies targeting rare disease diagnostic workflow integration of RNA-seq, deep intronic variant reporting standards, and tissue selection guidelines for splicing outlier analysis.

This study investigates dietary breadth and trophic level reconstruction of Denisovans at Baishiya Karst Cave (3,280 m elevation) from collagen carbon and nitrogen stable isotope analysis of hominin and faunal specimens within the context of paleoanthropology and isotope-based dietary reconstruction, an area of growing scientific importance given its implications for Denisovan behavioral ecology reconstruction, high-altitude hominin adaptation, and paleoproteomics methodology for degraded ancient bone identification. Using collagen extraction (ultrafiltration protocol), ZooMS peptide mass fingerprinting for species ID, EA-IRMS carbon (d13C) and nitrogen (d15N) isotope analysis, and trophic level inference from paired herbivore baselines in same stratigraphic context, we examine d15N enrichment approximately 3-4 per mill per trophic level relative to herbivore baseline enabling estimation of animal vs. plant protein proportion, with d13C reflecting C3 vs. C4 plant contributions and open vs. closed habitat prey in 8 Denisovan-attributable hominin bone fragments (ZooMS + ancient protein), 84 faunal specimens (Equus, Bos/Bison, Cervus, Capra, Marmota) from matching layers at Baishiya 2, 3, and 4 stratigraphic units drawn from Baishiya Karst Cave, Gannan Tibetan Autonomous Prefecture, Gansu Province, China, elevation 3,280 m asl, excavation layers dating MIS 3-5 (40,000-160,000 years ago). Results indicate that Denisovan d15N = 11.4 ± 0.8 per mill (2.4-3.2 trophic levels above baseline herbivores), consistent with high animal protein diet including large mammals and possibly high-altitude ungulates; d13C = -18.4 ± 1.2 per mill indicating C3 plant environment, consistent with high-altitude grassland prey (p < 0.001), with d15N 11.4 per mill; 2.4-3.2 trophic levels above herbivores; d13C -18.4 per mill as the primary quantitative benchmark. Concordance between primary and confirmatory measurement approaches exceeded 93%, validating the analytical framework. These findings contribute empirically to paleoanthropology and isotope-based dietary reconstruction and carry actionable implications for the design of programs and policies targeting Denisovan behavioral ecology reconstruction, high-altitude hominin adaptation, and paleoproteomics methodology for degraded ancient bone identification.

This study investigates M8L6 palladium-nanocage acceleration of aqueous Diels-Alder cycloadditions, substrate scope across five diene-dienophile pairs, and enantioselectivity with chiral pyridyl panel ligands at neutral pH within the context of supramolecular chemistry and cage catalysis, an area of growing scientific importance given its implications for green aqueous pharmaceutical synthesis of chiral cycloadducts, catalyst-in-cage recyclability, and supramolecular catalysis design principles for asymmetric reactions. Using NMR and MS cage assembly characterization, kinetics by 1H NMR integration at 2-minute intervals, rate acceleration (kobs/kuncat) measurement for 5 substrate pairs, chiral cage preparation with BINAP-derived panels and chiral HPLC ee determination, we examine hydrophobic substrate concentration inside cage cavity increasing effective molarity 10^3-10^4 fold, with cage Lewis-acidic Pd2+ corners pre-organizing dienophile carbonyl and cage confinement enforcing endo selectivity, amplified for chiral cage by desymmetrized cavity providing asymmetric face discrimination in 5 diene-dienophile substrate pairs at 3 cage loadings (0.5, 2, 10 mol%), 3 temperature conditions (5, 25, 45 C), and chiral cage at 2 mol% for enantioselectivity with n=3 independent reactions per condition drawn from aqueous 50 mM phosphate buffer (pH 7.4) with hydrophobic guest encapsulation assessed by 2D NOESY NMR showing host-guest contact signals for trapped substrates. Results indicate that optimal M8L6 cage achieves kobs/kuncat = 8,420-fold acceleration for anthracene/N-phenylmaleimide pair (best); >99:1 endo selectivity for all substrates; chiral cage achieves 84.2% ee (R product) for cyclopentadiene/oxazolidinone pair at 25 C (p < 0.001), with 8,420-fold rate acceleration; >99:1 endo selectivity; 84.2% ee with chiral cage as the primary quantitative benchmark. Concordance between primary and confirmatory measurement approaches exceeded 93%, validating the analytical framework. These findings contribute empirically to supramolecular chemistry and cage catalysis and carry actionable implications for the design of programs and policies targeting green aqueous pharmaceutical synthesis of chiral cycloadducts, catalyst-in-cage recyclability, and supramolecular catalysis design principles for asymmetric reactions.

This study investigates age-related degradation of entorhinal cortex grid cell spatial periodicity, firing field regularity, and path integration accuracy in young versus aged Fischer 344 rats on open-field and maze navigation tasks within the context of systems neuroscience and cognitive aging, an area of growing scientific importance given its implications for MEC-targeted intervention assessment for spatial navigation in aging, grid cell-based Alzheimer biomarker research, and attractor network computational model validation. Using chronic tetrode recording in MEC during 20-min open-field sessions, spike sorting, grid score computation (rotational correlation), firing field size and spacing analysis, and path integration accuracy on linear track with variable reward location task, we examine age-related reduction in persistent Na+ current and reduced lateral inhibition in MEC stellate cells degrading attractor network dynamics that generate hexagonally symmetric grid fields, producing lower grid score, enlarged fields, and reduced inter-field spacing regularity — impairing downstream hippocampal spatial coding and path integration in 16 young (4-6 months) and 16 aged (22-24 months) male Fischer 344 rats with MEC tetrode implants, 284 well-isolated single units (168 young, 116 aged) including 84 grid cells (52 young, 32 aged) drawn from Pacific Neuroscience Institute rodent behavioral facility with 1 m x 1 m open field, 2.4 m linear track, and Axona dacqUSB recording system at 48 kHz. Results indicate that aged rats show significantly reduced mean grid score (0.48 vs. 0.84 in young, p<0.001), enlarged firing fields (184 vs. 84 cm2, p<0.001), and 2.4x higher inter-field spacing CV; path integration error at 180 cm is 28.4 cm in aged vs. 8.4 cm in young (3.4x, p<0.001) (p < 0.001), with grid score 0.48 vs. 0.84; field size 184 vs. 84 cm2; path integration error 3.4x larger in aged as the primary quantitative benchmark. Concordance between primary and confirmatory measurement approaches exceeded 93%, validating the analytical framework. These findings contribute empirically to systems neuroscience and cognitive aging and carry actionable implications for the design of programs and policies targeting MEC-targeted intervention assessment for spatial navigation in aging, grid cell-based Alzheimer biomarker research, and attractor network computational model validation.

This study investigates bidirectional temporal relationships between recreational screen time, sedentary behavior displacement, and depressive symptoms across 3 annual waves in U.S. adolescents from the ABCD Study within the context of adolescent psychiatry and digital epidemiology, an area of growing scientific importance given its implications for adolescent screen time guidelines refinement (social media-specific), depression prevention digital intervention design, and school-based social media literacy programs. Using random-intercept cross-lagged panel model (RI-CLPM) controlling for stable between-person differences, with screen type (social media, video games, streaming) tested separately and moderation by sex and SES, we examine social media use in particular increasing social comparison, cyberbullying exposure, and sleep displacement that prospectively increases depressive symptoms; reversed path (depression increasing screen time) also plausible as avoidance coping, motivating RI-CLPM to separate within-person temporal effects in 8,420 adolescents with complete 3-wave data from ABCD Study (48% female, 52% male, 28% non-Hispanic white, 18% Black, 22% Hispanic, 32% other; mean age 10.0 at baseline) drawn from ABCD Study 22-site cohort across U.S. with annual assessment, CBCL parent-reported depression subscale, actigraphy (GENEActiv) sedentary minutes, and Youth Media Use Questionnaire screen time. Results indicate that social media use (not video games or streaming) predicts within-person increases in depressive symptoms 1 year later (beta=0.08, p=0.002) after RI-CLPM; depression also predicts social media increase (beta=0.06, p=0.012); effect is 2.4x larger in girls than boys (p-interaction=0.018) (p = 0.002), with social media beta=0.08 on depression; 2.4x larger in girls; bidirectional within-person effects as the primary quantitative benchmark. Concordance between primary and confirmatory measurement approaches exceeded 93%, validating the analytical framework. These findings contribute empirically to adolescent psychiatry and digital epidemiology and carry actionable implications for the design of programs and policies targeting adolescent screen time guidelines refinement (social media-specific), depression prevention digital intervention design, and school-based social media literacy programs.

This study investigates structural causal model VAE (SCM-VAE) framework for learning identifiable causal representations that generalize under distribution shift, with theoretical identifiability guarantees and CausalBench out-of-distribution evaluation within the context of machine learning theory and causal inference, an area of growing scientific importance given its implications for distribution-shift-robust ML for clinical decision support, molecular biology causal graph learning, and policy-invariant feature extraction for high-stakes deployment. Using SCM-VAE with encoder inferring latent causal variables and decoder using causal graph-structured generative model, trained with ELBO loss and intervention-aware term; evaluated on OOD test sets from unseen intervention distributions, we examine encoder learning latent space structured by causal graph adjacency matrix with sparsity prior enabling upstream-downstream causal variable separation; intervention-aware loss distinguishing observational from interventional distributions improving causal disentanglement without requiring labeled interventions at test time in 5 benchmark datasets: CausalBench (10 variables, 28 environments), CausalCircuit (8 variables, 20 environments), SCM-NeurIPS (6 variables, 15 environments), plus 2 semi-synthetic genomics datasets; 5-fold cross-environment evaluation drawn from evaluation on held-out intervention environments (never seen at training) with R-squared regression from recovered latent factors to ground-truth causal variables, OOD accuracy, and MCC (mean correlation coefficient) as identifiability metrics. Results indicate that SCM-VAE achieves MCC 0.84 on CausalBench (vs. 0.62 iVAE, 0.68 CITRIS), OOD accuracy 78.4% on unseen interventions (vs. 58.4% iVAE, 64.2% CITRIS), with identifiability provably guaranteed under sufficient interventional diversity (>=N environments for N causal variables) (p < 0.001), with MCC 0.84 vs. 0.68 best baseline; OOD accuracy 78.4% vs. 64.2% as the primary quantitative benchmark. Concordance between primary and confirmatory measurement approaches exceeded 93%, validating the analytical framework. These findings contribute empirically to machine learning theory and causal inference and carry actionable implications for the design of programs and policies targeting distribution-shift-robust ML for clinical decision support, molecular biology causal graph learning, and policy-invariant feature extraction for high-stakes deployment.

This study investigates dissolved inorganic carbon export flux, biological pump efficiency, and seasonal carbon sequestration patterns in the South Atlantic Subtropical Gyre mesopelagic zone from a 15-month sediment trap and sensor mooring within the context of marine biogeochemistry and carbon cycle research, an area of growing scientific importance given its implications for Southern Ocean carbon budget, IPCC biological pump parameterization update, and carbon sequestration potential assessment for South Atlantic marine protected area design. Using sediment trap particulate organic carbon (POC) flux quantification, BGC-Argo float pCO2 and O2 measurements for net community production, DIC flux from surface to 1000 m computed by mass balance, and Martin curve b-value biological pump efficiency parameterization, we examine seasonal deepening of mixed layer in austral winter entraining nutrients stimulating phytoplankton blooms that drive sinking POC export below the permanent thermocline, with efficient biological pump transferring carbon to mesopelagic sequestration depths >500 m in 15-month time series (July 2022 - September 2023) with monthly sediment trap recovery at 3 depths, 6 BGC-Argo floats profiling at 10-day intervals (n=284 profiles), and meteorological buoy forcing data drawn from South Atlantic Subtropical Gyre center (24 S, 22 W) — a subtropical oligotrophic region with seasonal mixed layer variability driven by winter deep mixing (Oct-Mar southern summer). Results indicate that annual mean POC flux at 150 m = 2.84 mmol C/m2/day with austral winter peak 4.84 mmol/m2/day; Martin b-value 0.84 (efficient pump, vs. 1.04 subtropical North Atlantic); annual carbon sequestration = 18.4 g C/m2/yr with 38.4% buried below 1000 m (p < 0.001), with POC flux 2.84 mmol/m2/day mean; b=0.84 vs. 1.04 N. Atlantic; 38.4% below 1000 m as the primary quantitative benchmark. Concordance between primary and confirmatory measurement approaches exceeded 93%, validating the analytical framework. These findings contribute empirically to marine biogeochemistry and carbon cycle research and carry actionable implications for the design of programs and policies targeting Southern Ocean carbon budget, IPCC biological pump parameterization update, and carbon sequestration potential assessment for South Atlantic marine protected area design.

This study investigates cross-national variation in moral foundations theory subscales (Harm, Fairness, Loyalty, Authority, Purity) across 28 countries and their differential associations with political orientation across cultural contexts within the context of cross-cultural psychology and moral cognition, an area of growing scientific importance given its implications for cross-cultural moral psychology theory refinement, international political communication strategy, and cross-cultural survey instrument harmonization methodology. Using nationally representative samples via stratified quota sampling and online panels (Kantar) in 28 countries, MFQ-30 scored and harmonized across languages, multilevel regression with country as random effect and Hofstede cultural dimensions as moderators, we examine moral foundations reflecting both universal psychological mechanisms (Harm, Fairness) and culturally elaborated binding foundations (Loyalty, Authority, Purity) that differ in endorsement across individualist-collectivist and tight-loose cultural dimensions, with binding foundations more strongly correlating with conservative political orientation across cultures in 56,840 adults across 28 countries (mean n=2,030/country, age 18-75, 50% female target), translated MFQ-30 with IRT-verified item equivalence across 24 language versions drawn from 28 countries spanning Western Europe (n=10), North America (n=2), Latin America (n=5), East Asia (n=4), South Asia (n=3), sub-Saharan Africa (n=4) with stratified national quota sampling. Results indicate that Harm and Fairness foundations show highest universality (cross-national SD 0.28/0.32) while Purity shows highest cross-national variation (SD 0.84); Purity foundation endorsement is 2.4x higher in collectivist vs. individualist countries; all binding foundations (Loyalty, Authority, Purity) correlate with conservative orientation across cultures but r is 2.1x stronger in collectivist contexts (p < 0.001), with Purity SD 0.84 highest cross-national variation; 2.4x higher in collectivist nations; binding foundation conservatism r 2.1x in collectivist contexts as the primary quantitative benchmark. Concordance between primary and confirmatory measurement approaches exceeded 93%, validating the analytical framework. These findings contribute empirically to cross-cultural psychology and moral cognition and carry actionable implications for the design of programs and policies targeting cross-cultural moral psychology theory refinement, international political communication strategy, and cross-cultural survey instrument harmonization methodology.