Uddalak Das

Rett syndrome (RTT) is an X-linked neurodevelopmental disorder arising predominantly from de novo loss-of-function mutations in MECP2, affecting approximately 1 in 10,000 live female births. MECP2 is a dosage-sensitive epigenetic regulator acting through its methyl-CpG-binding (MBD) and transcriptional-repression (TRD) domains; isoform-specific expression and X-chromosome-inactivation mosaicism drive the wide phenotypic variability, with mutation type broadly correlating with severity. Although historically viewed as neuron-centric, RTT is a multisystem disorder: beyond psychomotor regression, stereotypic hand movements, gait apraxia, and loss of spoken language, MECP2 deficiency contributes to brainstem-mediated breathing dysrhythmia, QTc prolongation, enteric dysmotility, low bone density, and mitochondrial/metabolic deficits, amplified by glial–neuronal crosstalk. Preclinical models, MECP2 rodents, patient iPSC-derived neurons, and cerebral organoids, reveal synaptic and dendritic deficits that improve substantially when MECP2 is reactivated in these systems, indicating that MECP2-deficient neurons remain viable rather than degenerating; whether comparable improvement can be achieved in affected individuals is not yet established. The therapeutic landscape now spans multidisciplinary supportive care, the first approved pharmacotherapy trofinetide, AAV-mediated MECP2 gene replacement (now in pivotal trials, with an important gene-therapy safety signal), antisense oligonucleotides for dosage normalisation, and emerging metabolic agents; validated biomarkers and objective outcome measures remain a major unmet need. Conclusion: RTT is a multisystem monogenic disorder whose management is shifting from symptom control toward disease modification. Phenotypic rescue after MECP2 restoration in rodent models provides the rationale for this shift but does not establish that RTT is reversible in patients; progress now depends on validated biomarkers, dose-controlled disease-modifying therapies, long-term outcome data, and equitable global access. Graphical abstract Integrated overview of Rett syndrome (RTT) from molecular cause to equitable translation. 1 MECP2 as a dosage-sensitive epigenetic reader encoded at Xq28, where a loss-of-function MECP2 variant impairs methylated-DNA binding and chromatin regulation and disrupts transcriptional, splicing, and synaptic gene networks; the MECP2-E1 (492 amino acids) and MECP2-E2 (486 amino acids) isoforms differ in distribution, and severity is modified by mutation type, residual function, and X-chromosome-inactivation mosaicism. Because too little MECP2 causes RTT while too much is also pathogenic, restoration must be dose-controlled, cell-aware, and brain-wide. 2 RTT extends well beyond the central nervous system, direct peripheral effects combining with brainstem autonomic dysfunction to involve brain and glia, breathing, cardiac, gastrointestinal, skeletal and metabolic systems, across the characteristic four-stage clinical trajectory. 3 A therapeutic landscape moving from whole-person supportive care toward controlled gene restoration, spanning trofinetide (modest symptom-level benefit), AAV gene replacement (milestone gains in uncontrolled single-arm programmes, with expression control and immune safety critical after a fatal high-dose event in 2024), and earlier-stage antisense, X-reactivation, editing, and small-molecule approaches. No approach has yet demonstrated reversal of established deficits in humans. 4 Translation, safety, and equitable access as the rate-limiting step, from earlier genetic diagnosis through registry-based natural-history data and biomarker-guided trials to equitable access What is Known: • More than 90–95% of classic Rett syndrome is caused by de novo loss-of-function mutations in MECP2, a dosage-sensitive epigenetic reader whose isoform distribution, residual function and X-chromosome-inactivation mosaicism together account for much of the phenotypic variability. • Reactivating MECP2 in symptomatic adult rodents substantially improves many disease features, establishing that MECP2-deficient neurons are functionally impaired rather than lost and providing the rationale for disease-modifying therapy. What is New: • This review integrates isoform-specific and non-neuronal (glial, respiratory, cardiac, gastrointestinal, skeletal and metabolic) mechanisms with a literature base updated to 15 August 2026, and identifies the continuing absence of any validated biomarker or surrogate endpoint, rather than target identification, as the principal obstacle to efficient trials. • It critically appraises the post-trofinetide landscape — divergent FDA and EMA assessments of the same trial data, a fatal high-dose AAV hyperinflammatory event, and single-arm registrational gene-therapy programmes reported largely outside peer review — and concludes that the demonstrated outcome to date is gain or re-acquisition of developmental milestones, not reversal of established disease.

Predictive multiscale cellular modeling is emerging as a consequential direction in precision medicine, converging hypothesis grammars, digital twins, and integrative genomics to interrogate tumor-immune dynamics, therapeutic resistance, and cellular plasticity. This perspective synthesizes recent progress across these domains and critically maps their translational potential alongside their current limitations. Hypothesis grammars translate mechanistic theories into executable agent-based models (ABMs) and hybrid ODE-PDE systems, enabling rapid in silico hypothesis testing while lowering the authoring barrier for domain scientists. Patient-specific digital twins, driven by multi-omics data, employ stochastic ensemble methods to simulate clonal evolution and microenvironmental interactions, though prospective clinical validation of these capabilities remains at an early stage. Integrative genomics, leveraging algorithms such as SCODE and SimiC, infers causal gene regulatory networks (GRNs) using Bayesian variational autoencoders, embedding dynamic intracellular logic into tissue-scale simulations. Emerging applications include in silico oncology trials for optimizing checkpoint blockade and combination therapies. Large language models are being explored to enhance rule induction, while FAIR-compliant digital cell repositories aim to ensure reproducibility and reuse. Verification, validation, and uncertainty quantification (VVUQ) via Sobol sensitivity analysis and Kennedy-O’Hagan calibration are identified as essential components for addressing non-identifiability and supporting regulatory credibility. Federated learning is discussed as a means of mitigating privacy and bias concerns in multi-institutional settings. Together, these converging approaches outline a plausible pathway toward virtual clinical trials and adaptive theranostics, contingent on the prospective validation, data infrastructure, and governance frameworks that clinical deployment will require. Graphical Abstract The alternative text for this image may have been generated using AI. Patient multi-omics data are processed through hypothesis grammars, digital twins, and integrative multiscale modeling into an AI-driven workflow that simulates tumor–immune dynamics, forecasts therapeutic outcomes, and iteratively guides precision oncology decisions under FAIR principles.

Rett syndrome (RTT), an X-linked neurodevelopmental disorder predominantly arising from de novo MECP2 mutations, manifests with psychomotor regression, stereotypic hand movements, gait apraxia, and expressive aphasia, driven by dosage-sensitive epigenetic dysregulation via MeCP2's methyl-CpG-binding domain (MBD) and transcriptional repression domain (TRD). Isoform-specific expression (MeCP2-E1 neuronal predominance) and X-chromosome inactivation mosaicism underpin phenotypic variability, with missense (R133C, T158M) and nonsense (R168X, R255X) variants correlating to severity gradients. Multisystem pathophysiology encompasses brainstem-mediated respiratory dysrhythmias, QTc prolongation via ion channel perturbations, enteric hypomotility, osteopenic fractures, and mitochondrial bioenergetic deficits, exacerbated by glial-neuronal crosstalk and oxidative stress. Preclinical platforms, including Mecp2-null rodents, patient-derived iPSCs, and cerebral organoids, elucidate synaptic hyperexcitability, dendritic arborization deficits, and reversibility upon Mecp2 reactivation. Therapeutic modalities span supportive multidisciplinary interventions, FDA-approved trofinetide (IGF-1 analog modulating neurotrophic cascades), AAV-mediated gene replacement (NGN-401, TSHA-102 with miRARE autoregulation), ASOs for dosage normalization, and emerging PPAR-γ agonists targeting metabolic homeostasis. Prioritized research agendas emphasize validated biomarkers (BDNF/IGF-1 axes, miRNA signatures), combinatorial regimens, and equitable global access to mitigate caregiver burden and phenotypic heterogeneity.

Autism Spectrum Disorder (ASD) is a genetically heterogeneous neurodevelopmental condition driven by rare de novo variants, copy number variations, and polygenic risk. SFARI-curated genes show high mutational constraint and enriched expression in cortical neurons and glia. This review highlights recent advances in CRISPR-based functional genomics using human pluripotent stem cells and induced pluripotent stem cells differentiated into neural progenitors, excitatory and inhibitory neurons, astrocytes, microglia, and brain organoids. CRISPR modalities including knockouts, CRISPRi and CRISPRa, base and prime editing, and Cas13 enable pooled and arrayed screens with high coverage at low multiplicity of infection. Integration of multimodal readouts such as Perturb-seq, single-cell and spatial transcriptomics, proximity labeling proteomics, and functional assays including microelectrode arrays and calcium imaging provides system-level insights into ASD gene function. Computational frameworks like MIMOSCA and SCEPTRE facilitate network reconstruction and pseudo-time inference. Case studies reveal Wnt and BAF complex dysregulation, microglial pruning deficits, and non-cell autonomous effects. Translational approaches target haplo-insufficient genes such as CHD8 and SCN2A using AAV or antisense oligonucleotides supported by isogenic iPSC models. Remaining challenges include model immaturity and scalability, while future directions focus on spatial perturb-omics, AI-driven causal inference, and standardized biobanks for precision ASD therapeutics.

Colorectal carcinoma (CRC) exerts a growing global disease burden, with microsatellite-stable/proficient mismatch repair (MSS/pMMR) tumors exhibiting intrinsic refractoriness to immune-checkpoint blockade (ICB) owing to low tumor mutational burden, limited neoantigenicity, and an immunosuppressive tumor microenvironment (TME) dominated by regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs). This review evaluates induced pluripotent stem cell (iPSC)–derived polyvalent vaccines as ontogenetically recapitulative immunogens capable of reinstating broad antitumor immunity. Reprogramming induces re-expression of oncofetal tumor-associated antigens, including cancer-testis antigens (NY-ESO-1, MAGE-A3), aberrant glycoforms of CEA and MUC1, and clinically actionable neoepitopes such as KRAS^G12D/V, thereby promoting epitope spreading and immunogenic cell death. Irradiated autologous or syngeneic iPSCs, delivered with Toll-like receptor 9 agonists, facilitate robust MHC I/II cross-presentation, driving CD8⁺ cytotoxic T-cell activation, Th1 polarization, perforin/granzyme-mediated cytolysis, and favorable effector-to-suppressor ratios. Preclinical models of melanoma, pancreatic ductal adenocarcinoma, and MSS CRC demonstrate prophylactic and therapeutic efficacy, with neoantigen-enhanced iPSCs synergizing with radiotherapy-induced DAMPs to achieve durable regressions and memory T-cell formation. Translational priorities include CRISPR-engineered hypoimmunogenic iPSC platforms, GMP-compatible non-integrating reprogramming, and combinatorial integration with STING agonists, ICB, CAR-NK cells, and LNP-mRNA constructs to enable biomarker-guided clinical deployment in minimal-residual-disease CRC.

Advances in single-cell and spatial multi-omics technologies have transformed the understanding of neutrophils from short-lived effector cells to highly heterogeneous and transcriptionally plastic immune populations. Within the inflamed intestinal microenvironment, gradients of cytokines, oxygen tension, and microbial metabolites such as short-chain fatty acids dynamically modulate neutrophil differentiation and function, shaping either tissue-protective or tissue-destructive phenotypes. Recent studies highlight the de novo expression of the NADPH oxidase enzyme DUOX2 in intestinal neutrophils as a pivotal mediator of redox signaling. DUOX2-derived reactive oxygen species activate epithelial and immune signaling cascades through NF-κB and p38 MAPK pathways, thereby amplifying inflammation, promoting barrier disruption, and sustaining microbial dysbiosis. Although this oxidative response enhances antimicrobial defense, it concurrently contributes to neutrophil extracellular trap (NET)-driven thrombo-inflammation and chronic tissue injury. Experimental evidence indicates that selective ablation of myeloid DUOX2 attenuates colitis, underscoring its potential as a therapeutic target. Emerging interventions that modulate this axis, including JAK/STAT inhibitors, CXCR2 antagonists, p38/MK2 inhibitors, and butyrate-based metabolic regulators, offer promising avenues to restore neutrophil homeostasis while maintaining host defense. Integrating single-cell transcriptomics, redox proteomics, and advanced imaging approaches will be essential for translating neutrophil plasticity into biomarker-guided and precision-based therapeutic strategies for durable mucosal healing in IBD.

Predictive multiscale cellular modeling redefines precision medicine by integrating hypothesis grammars, digital twins, and integrative genomics to forecast tumor–immune dynamics, therapeutic resistance, and cellular plasticity. Hypothesis grammars translate mechanistic theories into executable agent-based models (ABMs) and hybrid ODE–PDE systems, enabling rapid in silico hypothesis testing. Patient-specific digital twins, driven by multi-omics data, employ stochastic ensemble methods to simulate clonal evolution and microenvironmental interactions. Integrative genomics, leveraging algorithms like SCODE and SimiC, infers causal gene regulatory networks (GRNs) using Bayesian variational autoencoders, embedding dynamic intracellular logic into tissue-scale simulations. Applications include in silico oncology trials optimizing checkpoint blockade and combination therapies. Large language models enhance rule induction, while FAIR-compliant digital cell repositories ensure reproducibility. Verification, validation, and uncertainty quantification (VVUQ) via Sobol sensitivity and Kennedy–O’Hagan calibration address non-identifiability. Federated learning mitigates privacy and bias concerns. This framework enables virtual clinical trials and adaptive theranostics, transforming biological understanding into actionable, equitable precision medicine.

Leukemia encompasses heterogeneous hematologic malignancies characterized by dysregulated hematopoiesis, metabolic reprogramming, and therapy resistance, with relapse driven by persistent leukemic stem cells (LSCs). This comprehensive review delineates the pivotal role of the taurine–TauT (SLC6A6) axis as a metabolic vulnerability across acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), chronic myeloid leukemia (CML), and chronic lymphocytic leukemia (CLL). Taurine, a sulfur-containing β-amino acid, modulates redox homeostasis via taurine chloramine (TauCl) formation, stabilizes mitochondrial tRNA uridine modifications for oxidative phosphorylation (OXPHOS), and regulates calcium fluxes and apoptosis. SLC6A6-mediated taurine uptake is upregulated in leukemic blasts and LSCs, correlating with adverse cytogenetics (e.g., FLT3-ITD, TP53 mutations), chemoresistance to cytarabine, doxorubicin, methotrexate, vincristine, tyrosine kinase inhibitors (TKIs), and BCL2 inhibitors (venetoclax), and minimal residual disease persistence in hypoxic bone marrow niches. Transcriptomic analyses from TCGA, BeatAML, and TARGET cohorts substantiate SLC6A6 as a prognostic biomarker. Therapeutic strategies, including selective SLC6A6 inhibition and chemosensitization, exploit this dependency to amplify ROS-induced apoptosis and eradicate LSCs, potentially enhancing remission durability. Future directions emphasize structure-guided inhibitor development and integrative metabolomic profiling for precision oncology.

Alzheimer's disease (AD), the predominant dementia etiology, imposes a $1.5 trillion global burden in 2025, exacerbated by inexorable failures in amyloid-tau-centric monotherapies amid <1% trial success rates. This review synthesizes cross-disciplinary pathobiology underpinning oncology drug repurposing for AD, leveraging inverse epidemiological antagonism and convergent hallmarks: aberrant neuronal cell-cycle re-entry (CDK5/p25, ERK-MAPK), proteostasis overload (HSP90-UPS-autophagy nexus), mTORC1 hyperactivation impairing lysosomal flux, kynurenine pathway (IDO1/TDO)-driven immunometabolic decoupling, PARP1-mediated NAD+ parthanatos, maladaptive neuroinflammation (PD-1/PD-L1 checkpoints, TAM-microglial skewing), and senescence-associated secretory phenotype (SASP)-amplified gliosis. Computational nomination pipelines—transcriptomic signature reversal (CMap/LINCS), network proximity in multi-omic knowledge graphs, and ligand-based docking—prioritize pleiotropic oncology agents: c-Abl/Src TKIs (nilotinib, bosutinib) enhancing autophagic Aβ/tau clearance; multikinase modulators (masitinib) resolving mast cell-microglial dyshomeostasis; rapalogs restoring proteostasis; IDO1 antagonists normalizing astrocytic lactate shuttling; PARPi mitigating genomic instability; HDACi reinstating synaptic epigenomes; and senolytics (dasatinib+quercetin) purging senescent glia. Phase I/II trials evince biomarker modulation (↓CSF p-tau, amyloid-PET; ↑autophagic flux) with geriatric-adapted dosing, yet underscore translational hurdles: BBB penetration, off-target liabilities, and polypharmacy in APOE ε4-enriched cohorts. A precision roadmap advocates orthogonal combinations (e.g., TKI+IDO1i), delivery-first PBPK modeling, and biomarker-anchored adaptive platforms to recalibrate AD's networked trajectory toward disease-modifying efficacy.