<![CDATA[Phase 2 IRIS finds ML-004 misses social communication goals in autism, but notably reduces irritability in teens.]]>

Dynamic changes of gut microbiota during progression of three Alzheimer’s disease mice models

IntroductionAlzheimer’s disease (AD) is an age-related and progressive neurodegenerative disorder characterized by cognitive impairment and irreversible neuronal degeneration, affecting approximately 55 million individuals worldwide. Despite extensive research efforts, the underlying pathogenic mechanisms of AD remain incompletely understood, and effective therapeutic strategies for preventing or delaying disease progression are still lacking. Increasing evidence suggests that the microbiota-gut-brain axis plays an important role in neurodegenerative diseases, including AD. However, the dynamic alterations of gut microbiota during AD progression across different transgenic mouse models remain poorly characterized.MethodsIn the present study, we investigated age-dependent changes in gut microbiota composition in three commonly used AD mouse models, including APP/PS1, 3xTg, and 5xFAD mice, using 16S rRNA gene sequencing. Fecal samples were collected longitudinally at 2, 4, 6, and 8 months of age to evaluate microbial diversity, community structure, and differential bacterial taxa during aging and disease progression.ResultsOur results demonstrated distinct and model-dependent alterations in gut microbiota composition across different stages of AD progression. Significant changes in microbial diversity and bacterial community structure were observed among the three AD mouse models and wild-type controls. In particular, dynamic alterations in Verrucomicrobiota, Proteobacteria, and Actinobacteriota were consistently identified during aging in AD mice. In addition, β-diversity, Linear discriminant analysis effect size (LEfSe), and correlation network analyses further revealed differential microbial signatures associated with different AD mouse models and age stages.DiscussionOverall, our findings provide additional evidence that gut microbiota composition undergoes dynamic alterations during aging in multiple AD mouse models and may be associated with AD-related progression. This study may contribute to a better understanding of microbiota-associated changes during AD development and provide a basis for future mechanistic studies targeting the microbiota-gutbrain axis in AD.

Applications of machine learning algorithms to detect digital addiction: a meta-analysis

Digital addiction (DA) has emerged as a significant global concern, yet traditional diagnostic methods relying on self-report questionnaires face subjective bias and threshold inconsistencies. Recent advances in machine learning (ML) offer promising alternatives for automated DA detection. This study conducted a systematic meta-analysis of 64 eligible studies (75 independent datasets; N = 165,624), employing both single-group proportion and bivariate diagnostic test accuracy (DTA) models. The pooled classification accuracy was 0.87 (95% CI [0.85, 0.90]), and the DTA framework yielded a robust AUC of 0.92, with balanced sensitivity and specificity (both 0.86). Subgroup analyses showed high accuracy across subtypes, particularly for internet (0.90) and social media addiction (0.86). Accuracy was comparable between survey-based and physiological data, though physiological markers demonstrated superior specificity (0.90). These findings underscore the potential of ML-driven tools as scalable screening instruments while emphasizing the need for representative sampling and standardized diagnostic criteria to advance digital mental health practice.

Gender-specific symptom outcomes on cariprazine treatment: a 12-month naturalistic longitudinal follow-up study in schizophrenia

IntroductionA growing body of literature is focusing on third-generation antipsychotics and their unique characteristics, but few studies have examined gender as a crucial factor in response profiles. The present study aims to address this gap by analyzing the outcomes of 12-month naturalistic treatment with cariprazine to elucidate changes in specific psychopathological domains between men and women.MethodsThe present 12-month longitudinal naturalistic study involved a sample of individuals diagnosed with schizophrenia according to the DSM-5-TR treated with cariprazine at the outpatients’ psychiatric services of a major university and community hospitals in Italy. The assessments conducted included sociodemographic data, the Structured Clinical Interview for the DSM-5 (SCID-5), and the Positive and Negative Symptom Scale (PANSS) Total and Subscale scores, as well as the PANSS-derived Marder factors. The PANSS was administered at three time points: before starting the treatment with cariprazine (T0), after 6 months (T1), and after 12 months (T2).ResultsFifteen male and 17 female subjects were assessed at the three time points. The mean dose of cariprazine was 4.2 ± 1.3 mg for men and 4.0 ± 1.5 mg for women. Both genders exhibited improvements in all PANSS subscale symptoms after 6 and 12 months of cariprazine treatment compared to the baseline, with the only exception of the Uncontrolled hostility/excitement Marder factor among men. Progressive improvements through time points in symptom subscales were found in both sexes, reaching numerical differences in every PANSS subscale in both sexes at T2. Gender specifc response profiles emerged after 6 and 12 months of treatment in the PANSS subscales and items in men and women.DiscussionCariprazine exhibited significant efficacy in both sexes, with no significant differences between men and women despite a gender specific response profile emerged. Additional studies are needed to further investigate the efficacy profile and long-term outcome of cariprazine treatment by gender.

Long‑Range Gene Networks Uncover 641 New Schizophrenia‑Associated Genes

Schizophrenia’s genetic landscape just expanded dramatically. A new study in Nature Genetics identifies 641 previously unrecognized genes associated with schizophrenia, thanks to a modeling framework that captures how distant genetic variants regulate gene expression through co‑expression networks. The work reframes schizophrenia not as a collection of isolated genetic hits, but as a disorder shaped by long‑range regulatory relationships across the brain. The study is titled, “Co‑expression‑based models improve eQTL predictions for transcriptome‑wide association studies and highlight new schizophrenia‑associated genes.”

The research team, led by Giulio Pergola, PhD, at the Lieber Institute for Brain Development (LIBD), developed two trans‑aware predictive models—INGENE and MODULE—that quantify how variants far from a gene influence its expression through co‑regulated partners. Traditional transcriptome‑wide association studies (TWAS) focus almost exclusively on cis‑expression quantitative trait loci (ciseQTLs), variants within ±1 Mb of a gene. But as the paper noted, “Most transcriptome‑wide association approaches primarily model local (cis) genetic effects, leaving much of gene regulation unexplained.” By contrast, the new models incorporate distal (trans) regulatory effects, capturing regulatory relationships that behave more like social networks than neighborhood blocks.

Using RNA‑seq data from six human post‑mortem brain regions and genetic data from more than 102,000 individuals, the team integrated cis‑based predictors (CIS, EpiXcan) with their new trans‑based frameworks. The combined approach improved gene‑expression prediction for 18,744 genes, and when applied to Psychiatric Genomics Consortium (PGC3) datasets, it identified 766 schizophrenia‑associated genes, including 641 not previously detected by TWAS.

Pergola said the field has been “looking for the light under the lamppost, focusing only on genes close to disease‑associated DNA variants.” By illuminating long‑range interactions, he explained, “we’ve essentially turned on lights across the entire neighborhood, revealing how distant genetic variants coordinate to build the genetic basis of schizophrenia.”

The findings converge on pathways involved in glutamate signaling, neuronal communication, immune processes, and neurodevelopment—biological systems repeatedly implicated in psychiatric risk. MODULE‑derived trans‑single nucleotide polymorphisms (SNPs) showed particularly strong enrichment for schizophrenia‑associated variants, and many overlapped with cis‑eQTLs for transcription factors such as GATAD2A, RERE, IRF3, and SP4, all previously prioritized in schizophrenia GWAS.

Daniel Weinberger, MD, CEO and director of LIBD, emphasized the shift in perspective: “Schizophrenia risk isn’t just about individual genes acting one after another—it’s about how networks of genes work together. Understanding these coordinated genetic programs brings us closer to precision psychiatry.”

By demonstrating that trans‑regulatory architecture is both detectable and biologically meaningful, the study provides a roadmap for expanding TWAS beyond local effects. It also underscores the importance of integrating multi‑region brain transcriptomics with large‑scale genetic cohorts to reveal disease‑relevant regulatory relationships.

The post Long‑Range Gene Networks Uncover 641 New Schizophrenia‑Associated Genes appeared first on GEN – Genetic Engineering and Biotechnology News.

Gene Therapy Restores Brain Function and Behavior in Fragile X Syndrome

A University of California, Riverside-led research team has developed a gene therapy that restored production of a missing brain protein, corrected abnormalities in brain circuitry, and improved behavior in a mouse model of Fragile X syndrome (FXS). The study, published in the journal Molecular Therapy Nucleic Acids, tested an adeno-associated virus (AAV)-based therapy carrying a normal human version of the FMR1 gene to produce the Fragile X messenger ribonucleoprotein (FMRP) and found that early treatment normalized several measures of brain activity while improving social behavior, exploratory behavior, and cognitive flexibility.

“In a typical brain, FMRP acts like a brake or a volume control,” said senior author Iryna Ethell, PhD, a professor of biomedical sciences at the UC Riverside School of Medicine. “Without it, neural circuits become overactive and less efficient, which contributes to many of the developmental and behavioral challenges associated with FXS.”

FXS is the most common single-gene cause of autism spectrum disorder. According to the researchers, the disorder typically manifests from expansion of CGG repeats in the 5′ untranslated region of FMR1. The mutation causes methylation and silencing of the gene, leading to a major reduction or complete loss of FMRP, an RNA-binding protein that regulates numerous messenger RNAs involved in synapse formation, maturation, and function. Loss of the protein can lead to abnormal synaptic activity and increased cortical hyperexcitability.

FXS can produce sensory hypersensitivity, seizures, anxiety, intellectual disability, developmental delays, repetitive behaviors, and social communication difficulty. Current treatments for this syndrome don’t seek to cure it, rather they are aimed at managing the associated symptoms of anxiety, hyperactivity, irritability, aggression, depression, and seizures.

The therapy developed by the research team was designed to replace missing FMRP rather than repair the original mutation. To do this, the researchers used an AAV9 viral vector to deliver human FMR1 isoform 7, one of the most abundant forms of the protein found in the brain. The therapy was tested in newborn mice lacking FMRP via intracerebroventricular injections at either a low or high doses.

The work built on earlier research that explored the potential of AAV-mediated restoration of FMRP in rodent models. These prior studies used a range of viral serotypes, promoters, delivery routes, and FMRP isoforms and showed they could partially or completely correct specific biochemical, physiological, and behavioral abnormalities. The researchers noted that studies involving mouse and rat FMRP homologs had shown that restoring the protein could improve a range of Fragile X-related deficits.

The current study showed that high-dose treatment produced the strongest positive effects in the mouse models. Electroencephalography showed normalization of baseline gamma power, improvements in responses to sound, reduced background neural activity, and improved habituation to repeated auditory stimuli. The therapy also restored abnormal patterns of brain-wave coupling that have been associated with Fragile X-related dysfunction.

Behavioral testing showed that these improvements persisted into adulthood. Mice receiving the higher dose displayed normalized exploratory behavior, improved social preference, and better performance in probabilistic reversal learning, a measure of cognitive flexibility that requires adapting when previously rewarded behaviors stop producing rewards.

“Fragile X mice tend to persist with an old solution even after the rules change,” Ethell said. “After treatment, they became much better at adapting, performing similarly to mice with normal FMR1 function.”

The researchers noted that their work showed the importance of delivering at therapy for FXS early in its development. They said that widespread distribution of the potential new gene therapy throughout the brain was necessary to achieve a therapeutic benefit. There was a clear relationship between the proportion of neurons expressing the therapeutic gene and the degree of functional recovery, which indicated that restoring FMRP in a sufficient number of cortical cells is critical for correcting any behavioral deficits.

While a promising step, the investigators said that the work was a preclinical study and that future research will now focus on developing delivery methods that can of have broad distribution across the human brain. The team also believes their approach could have broader applications.

“Beyond FXS, the findings may provide a roadmap for treating other genetic neurodevelopmental disorders caused by the loss of a single critical protein,” Ethell said. “Our study shows it may be possible to restore function across complex brain networks by replacing a missing gene. That gives us reason to be optimistic about the future of genetic medicine.”

The post Gene Therapy Restores Brain Function and Behavior in Fragile X Syndrome appeared first on Inside Precision Medicine.

Functional Outcome Prediction in Young Adults With Mental Health Symptoms Using Machine Learning and Large Language Models: Longitudinal Observational Study

Background: Functional impairments associated with mental health conditions are on the rise. Predicting functional outcomes may improve the targeting of preventive interventions. While prognostic models have primarily focused on psychosis, early recognition services require a transdiagnostic approach. Objective: This study aimed to predict global functioning within a 2-year follow-up using baseline clinical and structural magnetic resonance imaging (MRI) data in a population-based sample of young, help-seeking individuals presenting with affective and anxiety symptoms as well as attention-deficit hyperactivity disorder. Methods: We classified 357 help-seeking individuals aged 18‐35 years recruited from 9 sites as “impaired” (Global Assessment of Functioning [GAF] ≤60; n=228) or “nonimpaired” (GAF>60; n=129) at year 1 and/or year 2 follow-up. GAF classification group status at follow-up was predicted using linear support vector machine (SVM), decision tree, and large language model (LLM) Llama-3 using clinical assessments and/or structural MRI. Leave-one-site-out (SVM) or external sample (LLM) was used for validation. Results: SVM achieved balanced accuracy of 69.2% using clinical features only. Items related to baseline occupational functioning, interpersonal relationships, cognitive functioning, psychotic and affective symptoms, as well as the presence of anxiety disorder, were most predictive. The decision tree further reduced the feature set to 5 predictive items, achieving balanced accuracy of 76.6%. Although amygdala and hippocampal subregions achieved balanced accuracy of 57.1%, structural MRI did not improve the overall prediction. Llama-3 performed comparably well to SVM (balanced accuracy of 72.6%). Conclusions: Machine learning demonstrated good performance in predicting global functioning. Interestingly, the out-of-the-box LLM performed comparably well without being trained or fine-tuned, highlighting the potential of leveraging free-text data for mental health prognosis.