Predicting a treatable huntington's disease patient / patient population and effects on disease progression for therapies targeting somatic instability

Abstract

Somatic instability (SI) in the mutant HTT allele is a high value therapeutic target for Huntington's Disease (HD). However, rates of SI and the speed of HD progression vary widely with inherited germline CAG repeat lengths across the HD patient population. Understanding this variability and projecting the treatable range of patients for SI-directed therapies is important to their design, preclinical development, and conferring optimal benefit to HD patients. A computational model and related method(s) track the percentage of healthy medium spiny neurons (MSNs) across the lifetime of HD patients. Changes in MSN survival are predicted based on an exemplary SI-directed interventional gene therapy. The range of treatable patients is also projected. The flexible framework provided by the computational model and related method(s) enables therapeutic performance prediction and projecting outcomes across the HD patient population and furthermore, serves as a valuable tool for defining the target populations for SI-directed therapies.

Description

086013-0589445PREDICTING A TREATABLE HUNTINGTON’S DISEASE PATIENT / PATIENT POPULATION AND EFFECTS ON DISEASE PROGRESSION FOR THERAPIES TARGETING SOMATIC INSTABILITYCross-Reference to Related Applications

[0001] This Application claims priority to U.S. Provisional Patent ApplicationNo. 63 / 762,551, filed on February724, 2025, and to U.S. Provisional Patent Application No. 63 / 931,438, filed on December 4, 2025, the entire content of which is incorporated herein by reference.Field of the Invention

[0002] The invention relates to computational models pertaining to the treatment of repeat expansion diseases and disorders, such as Huntington’s Disease.Introduction

[0003] Huntington’s Disease (HD) is a neurodegenerative disorder caused by an expansion of a cytosine-adenine-guanine (CAG) trinucleotide repeat in the Huntingtin (HTT) gene. While an individual inherits a pathogenic HTT allele of a defined repeat length from a parent, the CAG repeat sequence undergoes progressive somatic expansion in certain cell types, including medium spiny neurons (MSNs), over the course of a lifetime. The extent of somatic expansion is stochastic and influenced by repeat length, with longer repeats exhibiting greater instability (Handsaker et al., 2025).

[0004] Medium spiny neurons demonstrate resilience to HTT CAG repeat lengths up to approximately 150. However, expansions beyond this threshold (above about 150) result in cellular dysfunction and eventual neurodegeneration. Notably, MSNs that accumulate HTT CAG repeat lengths exceeding 300 are rapidly lost due to cell death, contributing to the progression of HD. Given the critical role of somatic instability in disease pathology, targeting this process has emerged as a promising therapeutic strategy for HD.

[0005] A comprehensive understanding of the distribution of healthy, dysfunctional, and degenerating MSNs within affected individuals facilitates assessing the potential impact of gene therapies designed to mitigate somatic expansion. The efficacy of such therapeutic interventions depends on the presence of a viable population of MSNs with HTT CAG repeat lengths within a treatable range. This underscores that early therapeutic intervention086013-0589445preserves functional neurons and slow disease progression. However, versions of these cells that can be studied are difficult to obtain.

[0006] Studies identified MSH3 variants associated with clinically relevant HD outcomes, including age at motor onset and rate of disease progression (9). MSH3, a key component of the mismatch repair (MMR) machinery, binds MSH2 to form the MutSp complex, which maintains DNA integrity. HD patients with variants having reduced MSH3 expression show slower disease progression. For example, an MSH3 polymorphism that results in -10% lower MSH3 expression reduced SI, slowed disease progression, and delayed disease onset by 1 year (8). Likewise, heterozygous loss-of-function mutations in MSH3 (i.e., 50% MSH3 expression at birth) delayed average age at motor onset by 10.6 years (10, 11). These findings provide compelling evidence that reductions in MSH3 expression levels, and even modest reductions, may meaningfully impact the clinical course for HD patients. Further supporting the link between MSH3 and SI, genetic and pharmacologic studies in mouse and human cell models demonstrate that MSH3 lowering prevents or slows SI (11-17). In MSNs derived from HD patient induced pluripotent stem cells (iPSCs), antisense oligonucleotide (ASO)-mediated knockdown (KD) of MSH3 by 41% proportionally reduced SI, while 83% KD roughly halted SI (11). Together, these studies establish MSH3 as a promising therapeutic target in HD; however, the quantitative relationship between SI lowering, MSN preservation, and predicted clinical benefit (e.g., years of delayed clinical progression) remains to be defined.

[0007] To develop an SI lowering therapeutic, a viral vector (AAV-DB-3) was engineered to express microRNAs (miRNAs) targeting MSH3 (denoted AAV-DB-3.miMSH3). Earlier studies showed that the AAV-DB-3 capsid variant robustly targets MSNs and the vulnerable cortical neuron population in nonhuman primates (NHPs) and mice, as well as transduces human iPSC-derived neurons (6, 18). AAV-DB-3.miMSH3 lowered MSH3 mRNA levels throughout the NHP caudate nucleus and putamen and significantly reduced SI in striatal tissues in HdhQl 11 mice, a murine model of HD. The model and supportive experimental data can be used to guide translation to patients.Summary

[0008] Somatic instability (SI) in the mutant HTT allele is a high value therapeutic target for HD. However, rates of SI and the speed of HD progression vary widely with inherited germline CAG repeat lengths across the HD patient population. Understanding this variability086013-0589445and projecting the treatable range of patients for Si-directed therapies is important to their design, preclinical development, and conferring optimal benefit to HD patients.

[0009] A computational model and related method(s) track the percentage of healthy neural cells, such as MSNs, across the lifetime of HD patients. This enables evaluation of SI lowering therapeutic approaches based on the degree of SI reduction and percentage of MSNs impacted. Changes in MSN survival are predicted based on a Si-directed interventional therapy, such as a gene therapy. The range of treatable patients is also projected. The flexible framework provided by the computational model and related method(s) enables therapeutic performance prediction across the HD patient population and serves as a valuable tool for defining a target patient or the patient populations for Si-directed therapies. Additionally, our model integrates natural history data across a spectrum of germline CAG repeat lengths and ages of intervention, enabling simulation of clinical outcomes for individual patients across the HD population and characterizing clinical translatability thresholds.

[0010] According to an embodiment, a method for modeling or projecting therapeutic effect on somatic instability and / or cell survival in Huntington’s Disease is provided. The method comprises encoding a plurality of cell objects with cell attributes and cell functions to represent neural cells, such as medium spiny neurons (MSN). The cell attributes comprise cell age, cytosine-adenine-guanine (CAG) repeat length, and an indication of treated or untreated treatment status. The cell functions are configured to facilitate simulated modification of individual cell objects. The method comprises instantiating a cell collection of the cell objects. The cell collection has a defined CAG repeat length. The cell collection is encoded with population functions configured to facilitate: a simulated cell lifetime of somatic expansion events for the CAG repeat length across the cell objects in a cell collection, and applying a therapeutic to the cell objects in the cell collection using a rate of cell transduction. The method comprises simulating, using the population functions and the cell functions, and based on the cell attributes and a defined CAG repeat expansion rate, lifecycles for the cell objects in the cell collection. The cell collection is configured to simulate Huntington’s disease progression under therapeutic treatment. The simulation comprises a life cycle of mutation events at the defined CAG repeat expansion rate across the cell objects in the cell collection. Mutation rates for cell objects receiving said therapeutic treatment are modified to reflect reduced CAG repeat expansion. The method comprises determining the therapeutic effect on somatic instability and / or cell survival in Huntington’s086013-0589445Disease based on the simulating, thereby modeling or projecting therapeutic effect on somatic instability and / or cell survival across the diverse HD patient population as characterized by a range of germline CAG lengths and ages of intervention.

[0011] In certain embodiments, the method comprises instantiating one, two or more cell collections, each having defined CAG repeat lengths, with at least one cell collection configured to simulate HD progression with therapeutic treatment and one cell collection without therapeutic treatment, such that the at least one cell collection is configured to function as a control.

[0012] In certain embodiments, determining the therapeutic effect on somatic instability and / or cell survival in HD based on the simulating comprises: identifying cell objects as healthy if their CAG repeat length remains below a threshold after the simulating, and comparing a quantity of healthy cell objects in the at least one cell collection configured to simulate HD progression without therapeutic treatment to a quantity of healthy cell objects in the cell collection configured to simulate HD progression under therapeutic treatment, at various time intervals during the lifecycles of the cell objects.

[0013] In certain embodiments, the threshold is between 50-200 CAG repeats, the threshold is 150 CAG repeats, or the threshold is 200 or more than 200 CAG repeats.

[0014] In certain embodiments, determining the therapeutic effect on somatic instability and / or cell survival in HD based on the simulating further comprises determining a therapeutic benefit for a treated population of cell objects as a temporal (time) delay in reaching a predefined HD milestone at one or more of the various time intervals compared to an untreated population of cell objects.

[0015] In certain embodiments, the therapeutic treatment is a gene therapy. In certain embodiments, the therapeutic treatment is a miRNA targeting MSH3 or PMS1, or overexpression of FAN1. In certain embodiments, the therapeutic treatment is a miRNA targeting MSH3. In certain embodiments, the therapeutic treatment is a miRNA targeting PMS1. In certain embodiments, the therapeutic treatment is a siRNA targeting MSH3 or PMS1. In certain embodiments, the therapeutic treatment is overexpression of FAN1. In certain embodiments, the therapeutic treatment is an AAV expressing FAN1. In certain embodiments, the therapeutic treatment is a siRNA targeting PMS1. In certain embodiments, the therapeutic treatment is an ASO targeting MSH3. In certain embodiments, the therapeutic086013-0589445treatment is an ASO targeting FAN1. In certain embodiments, the therapeutic treatment is an ASO targeting PMS1.

[0016] In certain embodiments, the therapeutic treatment is a non-gene therapy. In particular aspects, non-gene therapies include, proteins and small molecules such as those targeting MSH3, siRNAs targeting MSH3, miRNAs targeting MSH3, anti-sense oligonucleotides targeting MSH3, gene editing approaches such as Zinc finger protein and CRISPR / Cas based gene editing to reduce or silence MSH3 expression, epigenetic effectors to reduce or silence MSH3 expression, non-viral gene therapy approaches, and non-viral (lipid nanoparticle delivered) gene editing approaches.

[0017] In certain embodiments, the predefined disease milestones comprise onset of volumetric changes to caudate and putamen as measured by, for example, one or more of magnetic resonance imaging (MRI), elevated cerebrospinal fluid (CSF) neurofilament like protein (NFL), motor symptoms, and / or HD stages HD ISS 1, 2, and / or 3.

[0018] In certain embodiments, determining the therapeutic effect on somatic instability and / or cell survival in HD based on the simulating comprises determining CAG repeat length for each cell object in each population at time intervals, and optionally comparing CAG repeat lengths from the treated population of cell objects to CAG repeat lengths from the untreated population of cell objects at the time intervals.

[0019] In certain embodiments, the method comprises repeating the instantiating and the simulating for populations of cell objects with different CAG repeat lengths and ages of therapeutic intervention, and determining the therapeutic effect on somatic instability and / or cell survival in HD based on CAG repeat length and therapeutic intervention age combination.

[0020] In certain embodiments, repeating the instantiating and the simulating for populations of cell objects with different CAG repeat lengths and cell ages, and determining the therapeutic effect on somatic instability and / or cell survival in HD based on CAG repeat length and age at therapeutic intervention comprises defining a model parameter space comprising a range of CAG repeat lengths and a range of cell ages, and executing simulations across the parameter space to generate predicted therapeutic benefits based on CAG repeat length and age.086013-0589445

[0021] In certain embodiments, the method comprises determining a treatable individual patient or population of individuals based on their ages and CAG repeat lengths.

[0022] In certain embodiments, the treatable individual patient or population of individuals is determined for a clinical application or clinical trial of the therapeutic.

[0023] In certain embodiments, the method comprises generating a map visualization of therapeutic effect values as a function of CAG repeat length and cell age when the therapeutic was applied.

[0024] In certain embodiments, the method comprises repeating the instantiating and the simulating for cell collections of cell objects with CAG repeat lengths from 1-200 repeating units and cell ages from about 1-100 years. In certain embodiments, the method comprises repeating the instantiating and the simulating cell collections of cell objects with CAG repeat lengths from 36-60 repeating units and cell ages from about 5-100 years. In certain embodiments, the method comprises repeating the instantiating and the simulating cell collections of cell objects with CAG repeat lengths from 36-60 repeating units and cell ages from about 5-60 years.

[0025] In certain embodiments, the CAG repeat length expansion rate is user defined.

[0026] In certain embodiments, the CAG repeat length expansion rate changes based on the CAG repeat length, e.g., expansion increases with increasing CAG repeat length.

[0027] In certain embodiments, the CAG repeat length expansion rate is associated with disease progression milestones at known CAG repeat lengths and age combinations.

[0028] In certain embodiments, the cell functions comprise obtaining CAG repeat lengths, determining a CAG repeat length for each cell object, triggering modeled therapeutic transduction events or cell -targeting events, obtaining therapeutic transduction status, modelling mutating a repeat, and modelling mutating a repeat in context of a cell to which a therapy has been administered.

[0029] In certain embodiments, the population functions comprise modelling mutating all cells in a population, modelling administering a gene therapy, modelling mutating cells after therapy administration, running a lifecycle modelling algorithm for all cells in a population, and running a lifecycle modelling algorithm with modeled therapy.086013-0589445

[0030] In certain embodiments, cell ages, CAG repeat lengths, treatment statuses, population sizes, characteristics of the therapeutic, rate of transduction, and / or the MSN somatic expansion rate comprise user defined variables.

[0031] In certain embodiments, the plurality of cell objects comprises up to about 100, 1000, 1000000, or 10000000 cell objects.

[0032] In certain embodiments, the cell collection comprises 1, or 2, or up to about 100, 1000, 1000000, or 10000000 cells.

[0033] In certain embodiments, the cell collection represents a single human patient.

[0034] In certain embodiments, the cell collection represents two or more human patients.

[0035] In certain embodiments, the model parameter space further comprises atreated-fraction parameter and a knockdown parameter, and executing simulations across the parameter space comprises systematically varying the treated-fraction parameter and the knockdown parameter across defined ranges to generate predicted therapeutic benefit values as a function of both parameters.

[0036] In certain embodiments, the treated-fraction parameter is varied across a range from about 10% to about 100%, optionally in increments of about 1%, 5%, or 10%.

[0037] In certain embodiments, the knockdow n parameter is varied across a range from about 10% to about 100%, optionally in increments of about 1%, 5%, or 10%, and wherein knockdown is modeled as a reduction in expression of a target gene product and / or as a reduction in a modeled somatic repeat expansion rate.

[0038] In certain embodiments, executing simulations across the parameter space comprises generating a matrix, grid, or surface of predicted temporal delay values corresponding to combinations of treated-fraction and knockdown values.

[0039] In certain embodiments, the method further comprises determining, based on the simulations, a minimum treated-fraction value and a minimum knockdown value sufficient to produce a predicted therapeutic benefit comprising a temporal delay of greater than 5 years in reaching a predefined Huntington’s Disease milestone relative to an untreated / control population.086013-0589445

[0040] In certain embodiments, the minimum treated-fraction value is at least about 40%, 50%, 60%, 70%. 80%. or 90%, and / or the minimum knockdown value is at least about 40%.50%, 60%, 70%, 80%, or 90%.

[0041] In certain embodiments, the minimum treated-fraction value is at least about 50% and the minimum knockdown value is at least about 50%.

[0042] In certain embodiments, the predefined Huntington’s Disease milestone comprises onset of volumetric changes in the caudate and / or putamen, onset of motor symptoms, elevation of cerebrospinal fluid neurofilament light, and / or progression to a specified HD-ISS stage.

[0043] In certain embodiments, the method further comprises outputting the minimum treated-fraction value and the minimum knockdown value as a minimum target product profile (TPP), optionally in a visual format, report, data file, or electronic record.

[0044] In certain embodiments, the target gene product comprises MSH3, and knockdown of MSH3 is modeled as reducing a somatic CAG repeat expansion rate in treated medium spiny neurons.

[0045] In certain embodiments, the method further comprises evaluating a candidate therapeutic configuration by simulating, for the candidate therapeutic configuration, a treated-fraction value for the MSNs and a knockdown value for the target gene product, and determining whether the candidate therapeutic configuration is predicted to produce a temporal delay of greater than 5 years in reaching a predefined Huntington’s Disease milestone relative to an untreated / control population.

[0046] In certain embodiments, determining whether the candidate therapeutic configuration is predicted to produce the therapeutic benefit comprises determining that the treated-fraction value is at least about 50% and the knockdown value is at least about 50%.

[0047] In certain embodiments, the method further comprises selecting the candidate therapeutic configuration for further development, clinical application, and / or clinical trial enrollment when the candidate therapeutic configuration is predicted to produce the therapeutic benefit, and rejecting the candidate therapeutic configuration otherwise.086013-0589445

[0048] In certain embodiments, when the candidate therapeutic configuration is rejected, the method further comprises modifying one or more parameters of the candidate therapeutic configuration, including modifying a rate of cell transduction, treated-fraction parameter, knockdown parameter, dose, timing of administration, and / or other modeled therapeutic characteristic, and repeating the simulating and evaluating steps.

[0049] In certain embodiments, the candidate therapeutic configuration comprises a gene therapy targeting MSH3, including a siRNA, miRNA, antisense oligonucleotide, zinc finger protein, CRISPR-based system, or epigenetic effector configured to reduce or silence MSH3 expression.

[0050] In certain embodiments, the treated-fraction parameter is derived from a modeled rate of cell transduction and / or from a specified distribution of transduction events across a simulated MSN population.

[0051] According to other embodiments, methods more broadly focused on repeat expansion disorders are provided. For example, according to an embodiment, a method for modeling or projecting therapeutic effect on somatic instability and / or cell survival in repeat expansion disorders is provided. The method comprises encoding a plurality of cell objects with cell attributes and cell functions to represent human cells. The cell attributes comprise cell age, a repeat length, and an indication of treated or untreated treatment status. The cell functions are configured to facilitate simulated modification of individual cell objects. The method comprises instantiating a cell collection of the cell objects. The cell collection has a defined repeat length. The cell collection is encoded with population functions configured to facilitate: a simulated cell lifetime of somatic expansion events for repeat lengths across the cell objects in a cell collection; and applying a therapeutic to the cell objects in the cell collection using a rate of cell transduction. The method comprises simulating, using the population functions and the cell functions, and based on the cell attributes and a defined somatic expansion rate, lifecycles for the cell objects in the cell collection. The cell collection is configured to simulate disease progression under therapeutic treatment. The simulation comprises a life cycle of mutation events at the defined expansion rate across the cell objects in the cell collection. Mutation rates for cell objects receiving simulated therapeutic treatment are modified to reflect reduced somatic repeat expansion. The method comprises determining the therapeutic effect on somatic instability and / or cell survival based on the simulating,086013-0589445thereby modeling or projecting therapeutic effect on somatic instability and / or cell survival in repeat expansion disorders.

[0052] In certain embodiments, the repeat expansion disorder is a genetic neurodegenerative disorder. In certain embodiments, the repeat expansion disorder is Huntington's disease (HTT, CAG), Huntington’s disease-like 2 (JPH3. CTG / CAG), spinal and bulbar muscular atrophy or Kennedy disease (AR, CAG), dentatorubral-pallidoluysian atrophy (ATN1, CAG), spinocerebellar ataxia (SC A) type 1 (ATXN1, CAG), SC A type 2 (ATXN2, CAG), SCA type 3 / Machado-Joseph disease (ATXN3, CAG), SCA type 6 (CACNA1A. CAG), SCA type 7 (ATXN7, CAG), SCA type 8 (SCA8, CTG / CAG), SCA type 10 (SCAIO. ATTCT / ATTTT), SCA type 12 (SCA12, CAG / CTG), SCA type 17 (SCA17, CAG / CAA), SCA type 27B (FGF14, GAA), SCA type 31 (SCA31, TGGAA), SCA type 36 (SCA36, GGCCTG), SCA type 37 (SCA37, ATTTC / ATTTT), SCA type 51 (SCA51), my otonic dystrophy type 1 (DMPK, CTG), myotonic dystrophy type 2 (CNBP / ZNF9, CCTG), Fragile X syndrome (FMRI, CGG), fragile X-associated tremor / ataxia syndrome (FXTAS, CGG premutation), fragile X-associated primary ovarian insufficiency (FXPOI, CGG premutation), C9orf72-associated amyotrophic lateral sclerosis and frontotemporal dementia (C9orf72, GGGGCC), neuronal intranuclear inclusion disease (NOTCH2NLC, CGG), oculopharyngeal myopathy and leukoencephalopathy (OPML.CGG), oculopharyngodistal myopathy types 1-3 (OPDM1-3, CGG), cerebellar ataxia, neuropathy, and vestibular areflexia syndrome (CANVAS) (RFC 1, AAGGG), familial adult myoclonic epilepsy 1 (FAME 1), familial adult myoclonic epilepsy 2 (FAME2), familial adult myoclonic epilepsy 3 (FAME3), familial adult myoclonic epilepsy 4 (FAME4), familial adult myoclonic epilepsy (FAME6). familial adult myoclonic epilepsy 7 (FAME7), wherein any of the FAME is caused by pentanucleotide repeats such as TTTCA, TTTTA, ATTTC, and ATTTT in intronic regions, Fuchs endothelial comeal dystrophy (TCF4, CTG / CAG intronic repeat), autosomal dominant tubulointerstitial kidney disease (MUC1, mono-C frameshift repeat), X-linked intellectual developmental disorders / Partington syndrome (ARX, GCG / CGC repeats); panhypopituitarism with growth hormone deficiency (SOX3, GCN), holoprosencephaly type 5 (ZIC2, GCN); VACTERL association (ZIC3, GCC), hereditary sensory and autonomic neuropathy type VIII (PRDM12, GCC), Fragile XE syndrome (AFF2, CCG), a small-repeat disorder in PRNP, or a small-repeat disorder in MARCHF6.086013-0589445

[0053] According to other embodiments, a non-transitory computer readable medium having instructions thereon is provided. The instructions, when executed by a computer, cause the computer to perform operations for executing one or more of the methods or embodiments described herein.Brief Description of Drawings

[0054] FIG. 1 illustrates a method for modeling or projecting therapeutic effect on somatic instability' and / or cell survival in HD.

[0055] FIG. 2 is a step by step software model design description.

[0056] FIGS. 3A-3B illustrate modeling MSN CAG length and cell survival over a lifetime.

[0057] FIGS. 4A-B illustrate tracking a simulated percentage of healthy MSNs over a lifetime with treatment.

[0058] FIGS. 5A-5D illustrate measurement of therapeutic benefit as years of delay to HD landmark or HD-ISS stage.

[0059] FIG. 6 illustrates how a treatable range of HD patients varies with transduction efficiency and MSH3 knockdown.

[0060] FIG. 7 is a logical-architecture block diagram that illustrates a system including a computing engine and other components configured to facilitate modeling or projecting therapeutic effect on somatic instability and / or cell survival in HD.

[0061] FIG. 8 is a diagram that illustrates an exemplary computing system in accordance with embodiments of the present system.

[0062] FIG. 9 shows simulating somatic instability in HD patient MSNs, individuals, and populations.

[0063] FIGS. 10A-10E show target engagement and MSH3 lowering in NHP brain.

[0064] FIG. 11 shows dose-dependent lowering of MSH3 mRNA in striatal MSNs.

[0065] FIG. 12 shows clinical therapeutic benefit predicted at all doses evaluated in NHPs.086013-0589445

[0066] FIG. 13A-13D show dose-dependent slowing of somatic CAG repeat instability in HdhQlll mice.

[0067] FIGS. 14A-14E shows in vitro screening identifies potent artificial miRNAs targeting MSH3.

[0068] FIGS. 15A-15C are graphs showing miR-30 and miR-451 as scaffolds for miMSH3 mediated MSH3 reduction.

[0069] FIGS. 16A-16B are graphs showing favorable miRNA processing in HEK293 cells, mouse striatum, and NHP putamen.

[0070] FIG. 17 shows low detectable off-target gene expression in HEK293 cells transfected with miMSH3-06.

[0071] FIG. 18A-18B are graphs showing high AAV biodistribution in NHP brain with minimal detection in other tissues.

[0072] FIGS. 19A-19C are graphs showing MSH3 mRNA lowering in striatal MSNs.

[0073] FIG. 20 is a graph showing dose-dependent AAV-DB-3 biodistribution in HdhQl 11 mouse brain tissues with minimal detection in peripheral tissues.

[0074] FIGS. 21A-21E show that AAV-DB-3.miMSH3-ll reduced somatic instability in HdhQlll mice.

[0075] FIGS. 22A-22C show an exemplary selection of an optimal clinical trial population, showing that model outputs support therapeutic development.Detailed Description

[0076] Techniques (e.g., a computational model and related method(s)) for modeling or projecting therapeutic effect on SI and / or cell survival in genetic neurodegenerative disorders, with a focus on HD, are described. The present techniques provide a computational modeling approach to predict the therapeutic impact of somatic instability-targeting gene therapies for HD. Somatic expansion of CAG repeats in the HTT gene is a key driver of disease progression, with MSNs accumulating progressively longer repeats over a patient’s lifetime. These techniques are configured to track the percentage of healthy MSNs across different stages of HD and simulate the effects of an intervention, such as targeting the mismatch086013-0589445repair protein MSH3 (and / or other similar proteins), which has been implicated in CAG repeat expansion. By integrating patient-specific factors, including inherited HTT CAG repeat length and age at treatment, these techniques provide a predictive framework for identifying patients most likely to benefit from somatic instability -targeting therapeutics.

[0077] A key advantage of this computational approach is its ability to simulate long-term disease progression and quantify the potential delay in HD symptom onset following therapeutic intervention. The model predicts how varying levels of therapeutic transduction efficiency and MSH3 (for example) knockdown impact MSN survival, allowing for optimization of treatment parameters. Simulations demonstrate that MSNs with CAG repeat lengths of about 50, 80, 100. 150 or greater become dysfunctional, while those surpassing 300 undergo rapid cell death. By preserving functional MSNs through targeted suppression of somatic instability, these techniques offer a means to extend the window for therapeutic intervention, reduce neuronal loss, and delay clinical disease milestones.

[0078] Furthermore, these techniques generate patient population-level projections to inform clinical trial design and patient stratification. Using natural history data, these techniques identify a treatable individual HD patient and / or range of HD patients based on their germline HTT CAG repeat length and age at intervention, and / or other factors. Heatmap analyses illustrate that higher levels of MSH3 knockdown confer greater therapeutic benefit compared to increased transduction efficiency alone, emphasizing the benefit of effective somatic instability7suppression. The ability7to forecast patient-specific outcomes enhances precision medicine approaches for HD and ensures that therapeutic strategies are tailored to maximize clinical benefit.

[0079] FIG. 1 illustrates a method 100 for modeling or projecting therapeutic effect on SI and / or cell survival in genetic neurodegenerative disorders, with a focus on HD. In some embodiments, one or more operations of method 100 may be implemented in or by an electronic software model (e.g., a computational model), a computing system (e g., modelling engine 712 shown in FIG. 7 and / or computing system 800 shown in FIG. 8), and / or other components described herein. In some embodiments, method 100 comprises encoding (operation 102) a plurality of cell objects with cell attributes and cell functions to represent MSNs, instantiating (operation 104) a cell collection of cell objects, simulating (operation 106) lifecycles for the cell objects in the cell collection, determining (operation 108) repeat lengths, determining (operation 110) the therapeutic effect on SI and cell survival in genetic086013-0589445neurodegenerative disorders (such as HD) based on the simulating, and predicting (operation 112) a treatable population.

[0080] The operations of method 100 are also intended to be illustrative. In some embodiments, method 100 may be accomplished with one or more additional operations not described, and / or without one or more of the operations discussed. Additionally, the order in which the operations of method 100 are illustrated in FIG. 1 and described below is not intended to be limiting. In some embodiments, one or more portions of method 100 may be implemented in and / or controlled by one or more processing devices (e.g., a digital processor, an analog processor, a digital circuit designed to process information, an analog circuit designed to process information, a state machine, and / or other mechanisms for electronically processing information). The one or more processing devices may include one or more devices executing some or all of the operations of method 100 in response to (machine readable) instructions stored electronically on an electronic storage medium. The one or more processing devices may include one or more devices configured through hardware, firmware, and / or software to be specifically designed for execution of one or more of the operations of method 100 (e.g., see discussion related to FIGS. 7 and 8 below).

[0081] At operation 102, a plurality of cell objects are encoded with cell attributes and cell functions to represent medium spiny neurons (MSN). The plurality of cell objects may comprise up to about 100, 1000, 1000000, 10000000, or more cell objects. The cell attributes comprise cell age, a repeat length (e.g., a CAG repeat length), and an indication of treated or untreated treatment status. The cell functions are configured to facilitate simulated modification of individual cell objects. The cell functions comprise obtaining (CAG) repeat lengths, determining a (CAG) repeat length for each cell object, triggering modeled therapeutic transduction events, obtaining therapeutic transduction status, modelling mutating a repeat, modelling mutating a repeat in context of a cell to which a therapy has been administered, and / or other functions.

[0082] At operation 104, a cell collection of the cell objects is instantiated. The cell collection has a defined (CAG) repeat length. The cell collection is encoded with population functions configured to facilitate: a simulated cell lifetime of somatic expansion events for the (CAG) repeat length across the cell objects in a cell collection, applying a therapeutic to the cell objects in the cell collection using a rate of cell transduction, and / or other operations. The population functions may comprise modelling mutating all cells in a population,086013-0589445modelling administering a gene therapy, modelling mutating cells after therapy administration, running a lifecycle modelling algorithm for all cells in a population, running a lifecycle modelling algorithm with modeled therapy, and / or other population functions.

[0083] In certain embodiments, the method comprises instantiating two or more cell collections (e.g., 2. 3, 4, 5, 10, 100, 1000, 1000000, or more cell collections), each having defined or predetermined (CAG) repeat lengths, with at least one cell collection configured to simulate HD (for example) progression without therapeutic treatment, such that the at least one cell collection is configured to function as a control. In certain embodiments, an individual cell collection comprises up to about 100, 1000, 1000000, 10000000, or more cells.

[0084] Operation 106 comprises simulating (with the electronic software model and / or other computing device), using the population functions and the cell functions, and based on the cell attributes and a defined (CAG) repeat expansion rate, lifecycles for the cell objects in the cell collection. The (CAG) repeat length expansion rate may be user defined or predetermined. The (CAG) repeat length expansion rate changes based on the (CAG) repeat length. The (CAG) repeat length expansion rate is associated with disease progression milestones at known (CAG) repeat lengths and age combinations. The cell collection is configured to simulate a genetic neurodegenerative disorder (e.g., HD) progression under therapeutic treatment. The simulation comprises a life cycle of mutation events at the defined (CAG) repeat expansion rate across the cell objects in the cell collection. Mutation rates for cell objects receiving said therapeutic treatment are modified to reflect reduced (CAG) repeat expansion.

[0085] Operation 108 comprises determining (CAG) repeat lengths for each cell object in the cell collection for every' simulated time interval, e g., year. These are determined based on starting (CAG) repeat lengths, the (CAG) repeat length expansion rate, cell age. and / or other information.

[0086] Operation 110 comprises determining the therapeutic effect on SI and cell survival in the genetic neurodegenerative disorder (e.g., HD) based on the simulating, thereby- modeling or projecting therapeutic effect on SI and / or cell survival in the genetic neurodegenerative disorder (e.g., HD). In certain embodiments, determining the therapeutic effect on SI and cell survival (e.g., in HD) based on the simulating comprises: identifying cell086013-0589445objects as healthy if their (CAG) repeat length remains below a threshold after the simulating, and comparing a quantity of healthy cell objects in the at least one cell collection configured to simulate (HD) disease progression without therapeutic treatment to a quantity of healthy cell objects in the cell collection configured to simulate disease progression under therapeutic treatment, at various time intervals during the lifecycles of the cell objects. The threshold may be between 80-200 (CAG) repeats or 100-200 (CAG) repeats, for example. In some embodiments, the threshold is 50, 80, 100, 150, 100-150 or 200 or more than 200 CAG repeats.

[0087] In certain embodiments, determining the therapeutic effect on SI and cell survival in HD (as a specific example of a neurodegenerative disorder) based on the simulating further comprises determining a therapeutic benefit for a treated population of cell objects as a temporal delay in reaching a predefined HD milestone at one or more of the various time intervals compared to an untreated population of cell objects. These predefined disease milestones may comprise onset of volumetric changes to caudate and putamen as measured by, for example, magnetic resonance imaging (MRI), elevated cerebrospinal fluid (CSF) neurofilament like protein (NFL), motor symptoms, HD stages HD ISS 1, 2, and / or 3, and / or other milestones.

[0088] In certain embodiments, determining the therapeutic effect on somatic instability and cell survival in HD (keeping with this example) based on the simulating comprises determining CAG repeat length for each cell object in each population at time intervals, and comparing CAG repeat lengths from the treated population of cell objects to CAG repeat lengths from the untreated population of cell objects at the time intervals.

[0089] In certain embodiments, method 100 may comprises repeating operations 106-110 (e.g., the instantiating and the simulating) for populations of cell objects with different (CAG) repeat lengths and ages of therapeutic intervention. This may include repeating the instantiating and the simulating for cell collections of cell objects with CAG repeat lengths from 36-60 repeating units and cell ages from about 5-60 years, for example.

[0090] Operation 110 may include determining the therapeutic effect on SI and cell survival (e.g., in HD) based on (CAG) repeat length and therapeutic intervention age combination. Repeating the instantiating and the simulating for populations of cell objects with different (CAG) repeat lengths and cell ages, and determining the therapeutic effect on086013-0589445somatic instability and cell survival (e.g., in HD) based on (CAG) repeat length and age at therapeutic intervention comprises defining a model parameter space comprising a range of (CAG) repeat lengths and a range of cell ages, and executing simulations across the parameter space to generate predicted therapeutic benefits based on (CAG) repeat length and age. Operation 110 may comprise generating a map visualization of therapeutic effect values as a function of (CAG) repeat length and cell age when the therapeutic was applied.

[0091] Operation 112 comprises determining a treatable population of individuals based on their ages and (CAG) repeat lengths, and the output from the simulation(s) (e.g., the heatmap). The treatable population of individuals may be determined for a clinical application or clinical trial of the therapeutic, for example.

[0092] In certain embodiments, cell ages, (CAG) repeat lengths, treatment statuses, population sizes, characteristics of the therapeutic, rate of transduction, the MSN somatic expansion rate, and / or other variables that are a part of method 100 comprise user defined variables. A user may enter, select, or otherwise define such variables using a computing system (e.g., see description of FIGS. 7 and 8 below) and / or other devices.

[0093] FIG. 2 is an example step by step software model design description. FIG. 2 illustrates six primary steps, which are each described below. In this example, therapeutic variables include 1. the level of SI reduction (example level of MSH3 knockdow n). 2. transduction rate in MSNs, 3. age at therapeutic intervention, and 4. germline HD CAG repeat length. Disease variables comprise the rate of CAG repeat expansion (Handsaker et al., two phase linear model).

[0094] Step 1 : Simulating the trajectory of a disease like HD that preferentially targets a specific cell type (e.g., neural cells such as MSNs) requires simulating disease progression at the level of individual cells. Thus the model utilizes an object-oriented programming approach, defining cell objects that represent individual MSNs. Each cell object is encoded with attributes, including cell age, CAG repeat length, and therapeutic status, as well as functions that facilitate simulated modifications of the CAG repeat length, either through natural somatic expansion or therapeutic intervention. The cell functions include retrieving and setting repeat lengths, initiating a therapeutic transduction event, and simulating mutation events with and without therapy. Every instance of an object of class cell has the listed attributes and can be modified or queried with the listed functions.086013-0589445

[0095] Step 2: In order to model disease progression across a population of MSNs, the computational model further defines a cell collection object, comprising a set of individual cell objects with a user-defined germline CAG repeat length. The population functions enable the simulation of CAG repeat expansion events over time, including the administration of gene therapy and its effect on mutation rates. Since each MSN is exposed to unique DNA repair slippage events that cause somatic CAG expansion, the model accounts for stochastic variability while allowing controlled application of a therapeutic intervention. A key advantage of this framework is the ability to directly compare disease trajectory in untreated versus treated cell collections. By assigning a group of cell objects to a cell collection object, the model can simulate a lifetime of stochastic somatic expansion events across all cells in a population. The model can also control stochastic application of a therapeutic to this population and using a defined rate of transduction apply the therapeutic to a set percentage of cells in the population.

[0096] Step 3: To enable such comparisons, the model instantiates two (in this particular example) separate populations of MSNs, one for simulating therapeutic intervention, where cells are defined as having received therapy based on a user-defined transduction rate parameter, and the other separate population for simulating natural disease progression. The model is configured to reflect the modified trajectory of a treated population of cells. To enable this assessment of therapeutic impact, the model creates population(s) of a user-defined number of cells with a user defined germline CAG repeat length. One population is used for simulating a therapeutic treated population- in which cells are defined as having received the therapeutic based on a user defined transduction rate parameter and subsequent modification of their therapeutic status attribute value. An (optional) second population is used for simulating natural disease progression, e.g., without treatment.

[0097] Step 4: Both (again in this example there are two) cell collections are subjected to a simulated lifecycle spanning the progression of HD. In the untreated population, somatic expansion events occur at a predefined CAG repeat expansion rate, leading to progressive neuronal dysfunction and cell loss over time. In contrast, the treated population undergoes a simulated therapeutic event, wherein a subset of cells is designated as transduced. The CAG repeat expansion rate for transduced cells is modified to reflect a reduced rate of somatic expansion, capturing the therapeutic impact on cellular stability. Mutation events are applied at their defined rates across a simulated lifetime. In the case of the therapeutically treated086013-0589445population this process includes a simulated therapeutic treatment event during which a subset of cells are designated as transduced and thereafter experience a modified rate of somatic CAG repeat expansion.

[0098] Step 5: The model tracks the percentage of healthy cells (e.g., CAG repeat length < 150) at each simulated year and compares the trajectory of disease progression between the treated and untreated populations. The therapeutic benefit is quantified in terms of years of delay to critical disease milestones, including the onset of motor symptoms and progression through HD-ISS stages 1, 2, and 3, for example. A notable innovation of this approach is the integration of longitudinal disease progression datasets, such as those from Enroll-HD, which provide known associations between germline CAG lengths, patient age, and disease stage transitions.

[0099] Step 6: To assess the broader applicability of the therapeutic approach, the model executes simulations across a wide parameter space, varying germline CAG repeat length (e g., 36-60) and treatment age (e g., 5-60 years). By iterating across these parameters, the model can generate heatmap visualizations that delineate the treatable range of HD patients, based on the expected years of therapeutic benefit. An advantage of this modeling framework is its ability’ to identify patient populations with a short time to detectable therapeutic effect, thereby refining clinical trial enrollment criteria. Additionally, the model facilitates optimization of gene therapy design by simulating the impact of different transduction efficiencies and MSH3 knockdown (for example) levels. Notably, results indicate (see below) that knockdown efficiency has a greater impact on therapeutic benefit than transduction rate alone, underscoring the importance of robust somatic instability suppression.

[0100] The model may incorporate Steps 1-5 into a single function and run that function across an entire parameter space of possible germline CAG repeat lengths and treatment ages. Each individual CAG repeat and age combination are then run through the model function and with therapeutic benefit as an output. The heatmap visualization of therapeutic benefit values (years of delay to onset of X) reveals the treatable range of HD patients. A similar approach can be used to output other variables like time to detectable change in a parameter of interest. This is useful for identifying patient populations with short time to detection of clinical benefit, the cross section of the patient population that would be ideal to enroll in a clinical trial, for example.086013-0589445DefinitionsClaim Construction and Interpretation

[0101] The terms ‘"comprises”, “comprising”, “containing”, “having” and the like can have the meaning ascribed to them in U.S. Patent law and can mean "includes." "including," and the like. The term “consisting essentially of’ likewise has the meaning ascribed in U.S. Patent law and is open-ended, allowing for the presence of more than that which is recited so long as the basic or novel characteristics of that which is recited are not changed by the presence of more than that which is recited, but excluding prior art embodiments.

[0102] The terms “a”, “and,” and “the” include plural referents unless the context clearly indicates otherwise. Thus, for example, reference to “a vector” includes a plurality of such vectors.

[0103] The term “about” means a range of values including the specified value, which a person of ordinary skill in the art w ould consider reasonably similar to the specified value. In embodiments, "‘about” means a range extending to ±10% of the specified value. In embodiments, “about” includes the specified value and “about” at the beginning of a string of values modifies each of the values by ±10%.

[0104] The term “and / or” means any one of the recited elements alone, or any combination of two or more of the recited elements, unless the context clearly dictates otherwise.

[0105] The term “or” is used in an inclusive sense (i.e., “and / or”) unless the context clearly dictates otherwise.

[0106] The term “plurality” means two or more.

[0107] The term “configured to” when used in the context of a device, system, software, data structure, function, or instruction set, refers to being designed, arranged, programmed, encoded, constructed, or otherwise set up to perform a recited function or to achieve a recited result. A structure or element that is "configured to" perform a function need not be performing the function at a particular time.

[0108] The terms “user defined” and “user-defined" refer to a parameter, value, setting, selection, or input that can be specified, selected, or adjusted by a user, operator, clinician, researcher, or other entity interacting with the system or method.086013-0589445

[0109] All numerical values or numerical ranges include integers within such ranges and fractions of the values or the integers within ranges unless the context clearly indicates otherwise. Thus, to illustrate, reference to reduction of 95% or more includes 95%, 96%, 97%, 98%, 99%, 100% etc., as well as 95.1%, 95.2%, 95.3%, 95.4%, 95.5%, etc., 96.1%, 96.2%, 96.3%, 96.4%, 96.5%, etc., and so forth. Thus, to also illustrate, reference to a numerical range, such as “1-4” includes 2. 3, as well as 1.1, 1.2, 1.3, 1.4, etc., and so forth. For example. “1 to 4 weeks” includes 7. 8, 9, 10, 11, 12, 13, 14, 15. 16. 17. 18. 19. 20, 21, 22, 23, 24, 25, 26, 27, or 28 days.

[0110] Further, reference to a numerical range, such as “0.01 to 10” includes 0.011, 0.012, 0.013, etc., as well as 9.5, 9.6, 9.7. 9.8, 9.9, etc., and so forth. For example, a dosage of about “0.01 mg / kg to about 10 mg / kg” body weight of a subject includes 0.011 mg / kg, 0.012 mg / kg, 0.013 mg / kg, 0.014 mg / kg, 0.015 mg / kg etc., as well as 9.5 mg / kg, 9.6 mg / kg, 9.7 mg / kg, 9.8 mg / kg, 9.9 mg / kg etc., and so forth.

[0111] Reference to an integer with more (greater) or less than includes any number greater or less than the reference number, respectively. Thus, for example, reference to more than 2 includes 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, etc., and so forth. For example, administration of a therapeutic “two or more” times includes 3, 4, 5, 6, 7. 8, 9, 10, 11, 12, 13, 14, 15, or more times.

[0112] Further, reference to a numerical range, such as “1 to 100” includes 1.1, 1.2, 1.3, 1.4, 1.5, etc., as well as 81, 82. 83. 84. 85, etc., and so forth. For example, “between about 1 minute to about 90 days” includes 1.1 minutes, 1.2 minutes, 1.3 minutes, 1.4 minutes, 1.5 minutes, etc., as well as one day, 2 days, 3 days, 4 days, 5 days .... 81 days, 82 days, 83 days, 84 days, 85 days, etc., and so forth.Modeling, Simulation, and Outputs

[0113] The term “modeling” refers to generating, using one or more computational operations, a representation of one or more biological processes, parameters, or outcomes. Modeling can include generating predicted values and / or time courses for one or more variables (e.g., repeat length, cell survival, disease milestones) based on one or more inputs and one or more defined rules, distributions, or algorithms.086013-0589445

[0114] The term “projecting” refers to estimating or predicting one or more values or outcomes that may occur at a future simulated time, including extrapolating from one or more model assumptions, parameters, and / or previously simulated states.

[0115] The terms “simulation” and “simulating” refer to executing one or more computational algorithms that approximate, emulate, or represent the evolution of a modeled system over simulated time, including generating changes to one or more attributes of one or more modeled entities (e.g., cell objects) based on defined rules, rates, probabilities, or distributions.

[0116] The terms “lifecycle” and “life cycle” refer to a sequence of simulated states and / or events for a cell object over simulated time, optionally beginning at a defined initialization time (e.g., a defined cell age) and ending at a defined termination time, a modeled cell death event, and / or a modeled disease milestone. A life cycle can include one or more mutation events, therapy transduction events, and / or survival determinations.

[0117] The term “time interval” refers to a simulated time point or simulated time window at which one or more outputs are determined, stored, displayed, compared, or otherwise evaluated. Time intervals can be uniform or non-uniform, discrete or continuous.

[0118] The term “model parameter space” refers to a set of parameter values (e.g., ranges of repeat lengths and ages) over which one or more simulations are executed, optionally in a systematic or grid-like manner, to generate one or more outputs (e.g.. predicted therapeutic benefits) as a function of the parameters.

[0119] The term “map visualization” refers to a visual representation (e.g., a heat map, contour map. surface plot, or other graphical depiction) of one or more output values across two or more input dimensions, such as therapeutic effect values as a function of repeat length and age at therapeutic intervention.

[0120] The term “outputting” refers to generating, storing, saving, writing, transmitting, displaying, and / or otherwise providing one or more values or results, including values for a minimum treated-fraction, minimum knockdow n, a minimum target product profile, predicted therapeutic benefits, and / or any intermediate or final simulation outputs.086013-0589445Computational Structures and Execution

[0121] The term "encoding” refers to representing, storing, organizing, and / or associating information in a computer-readable format, including in memory, in a data structure, and / or in executable instructions. Encoding a cell object with attributes and functions can include creating data fields for the attributes and associating routines, methods, or callable operations for the functions.

[0122] The term “instantiating” refers to creating or establishing an instance of a data structure (e.g., a cell object or a cell collection) within a modeling framework, optionally with initialized values for one or more attributes. Instantiating can be employed via, but is not limited to any particular processing, memory architecture or storage mechanism.

[0123] The term “cell object” refers to a data structure, software object, record, class instance, or other computational representation of a cell, wherein the representation is associated with one or more attributes (e.g., repeat length, age, treatment status) and optionally one or more functions for simulating modification of the representation.

[0124] The term “cell attributes” refers to one or more variables, parameters, states, flags, values, or other data associated with a cell object. Cell attnbutes can be static or dynamic and can be updated during a simulation.

[0125] The term “cell functions” refers to one or more routines, methods, callable operations, or instruction sets associated with a cell object that, when executed, perform one or more simulated operations on the cell object and / or its attributes, including obtaining or updating repeat length, modeling mutation events, and / or updating treatment status.

[0126] The term “cell collection” refers to a set, collection, list, array, or other grouping of cell objects. In embodiments, a cell collection comprises one or more biological populations of cells (e.g., from one or more tissue, region, or patient). In embodiments, a cell collection represents a single tissue, region, or human patient. In embodiments, a cell collection represents a single human patient.

[0127] The term “population functions” refers to one or more routines, methods, callable operations, or instruction sets that act on two or more cell objects and / or an entire cell collection, including applying population-level rules, executing a lifecycle modeling086013-0589445algorithm, administering a modeled therapy to a subset of the population, and / or updating cell objects according to a defined expansion rate.

[0128] The term “operations” refers to one or more steps, functions, instructions, procedures, and / or computational acts performed by a computer when executing instructions, including the acts recited in the claims.

[0129] The term “instructions” refers to computer-executable code, software, firmware, and / or other machine-readable commands that, when executed by a computer or processor, cause the computer or processor to perform one or more operations.

[0130] The term “computer” refers to one or more processors and associated memory and / or hardware configured to execute instructions, including a server, workstation, personal computer, laptop, tablet, mobile device, cluster, or cloud computing resource.

[0131] The term “non-transitory computer readable medium” refers to a tangible computer-readable storage medium that stores computer-executable instructions, such as a memory, disk, solid-state storage, or other physical storage device. “Non-transitory” excludes a transitory, propagating signal per se.Biological Entities, Genetics, and Repeat Dynamics

[0132] The term “cell” refers to a biological cell unless the context clearly indicates that a computational representation (e.g., a cell object) is intended.

[0133] The term “medium spiny neuron (MSN)” refers to a neuronal cell type found in the striatum (including the caudate nucleus and putamen). In the context of a simulation, an MSN can be represented by a cell object that is configured to model one or more MSN-relevant attributes or behaviors, including somatic CAG repeat expansion.

[0134] The term “cell age” refers to an age value associated with a cell object, which can represent (i) a chronological age of a corresponding individual or patient, (ii) a time since a cell’s birth, differentiation, or other reference event, and / or (iii) a simulated time coordinate. Cell age can be expressed in years or other time units.

[0135] The term “age of therapeutic intervention / age at therapeutic intervention” refers to an age value (e.g., patient age or modeled cell age) at which a therapeutic is applied or is modeled as being applied in the simulation. In embodiments, age at therapeutic intervention086013-0589445is a simulation input parameter used to initialize treatment status and / or to schedule modeled transduction events.

[0136] The terms "cytosine-adenine-guamne (CAG) repeat” and “CAG / polyglutamine repeat” refer to a trinucleotide repeat comprising the nucleotide sequence CAG repeated in tandem. CAG encodes glutamine, a CAG repeat encodes polyglutamine.

[0137] The terms “non-CAG repeat” and “non-CAG motif’ refer to a nucleotide repeat comprising a nucleotide sequence other than CAG repeated in tandem. In some embodiments the non-CAG repeat is a trinucleotide repeat. In some embodiments, the non-CAG repeat is up to 12 nucleotides.

[0138] The term “repeat unit / repeating unit” refers to a repeating sequence motif whose tandem repetition is counted to determine a repeat length (e.g., the trinucleotide “CAG” as a repeat unit for CAG repeat length).

[0139] The term “CAG repeat length” refers to a number of CAG repeating units (repeat count) at a specified locus, including an expanded repeat associated with Huntington’s Disease. In embodiments, CAG repeat length is tracked per cell object and can change over simulated time due to somatic expansion events.

[0140] The term “repeat length” refers to a number of repeating units of a repeat sequence (e.g.. a trinucleotide repeat, tetranucleotide repeat, or other repeat motif) at a specified locus, including repeat lengths associated with genetic neurodegenerative disorders.

[0141] The term “somatic” refers to non-germline cells, tissues, and / or changes occurring within an individual after conception. Somatic changes can vary among cells within the same individual.

[0142] The terms “somatic expansion” and “somatic repeat expansion” refer to an increase in repeat length in one or more somatic cells over time, including in the context of CAG repeats. In embodiments, somatic expansion encompasses a distribution of repeat lengths across a population of cells.

[0143] The term “somatic expansion event” refers to an event in which a repeat length in a cell object changes in the simulation, including an increase in repeat length and, in086013-0589445embodiments, a decrease in repeat length. A somatic expansion event can be treated as a type of mutation event.

[0144] The term “somatic instability ” refers to the propensity7for, or the occurrence of, changes in repeat length in somatic cells over time, including expansions and, in embodiments, contractions, and resulting in a mosaic distribution of repeat lengths across cells and / or tissues.

[0145] The terms “somatic expansion rate” and “CAG repeat expansion rate” refer to a rate parameter describing how repeat length changes over time in a simulation, including an expected number of repeat changes per unit time, a probability distribution of repeat changes, and / or a function dependent on repeat length and / or age. In embodiments, the expansion rate is user defined and / or changes based on repeat length.

[0146] The term “mutation event” refers to a simulated event that changes one or more attributes of a cell object, including a change in repeat length (e.g., a somatic expansion event) and / or a modeled cell state change associated with disease progression.Disease Context, Comparators, and Milestones

[0147] The terms “repeat expansion disease” and “repeat expansion disorder” refer to genetic disorders associated with expansion of nucleotide repeats, and characterized by¬ progressive disease progression. In some embodiments, the repeat expansion disease primarily affects the central nervous system. Non-limiting examples of repeat expansion diseases and disorders include: Huntington’s disease (HTT, CAG), Huntington’s disease-like 2 (JPH3, CTG / CAG), spinal and bulbar muscular atrophy or Kennedy disease (AR, CAG), dentatorubral-pallidoluysian atrophy (ATN1, CAG), the spinocerebellar ataxias (SCAs) including type 1 (ATXN1, CAG), type 2 (ATXN2, CAG), type 3 / Machado-Joseph disease (ATXN3, CAG), type 6 (CACNA1A, CAG), type 7 (ATXN7, CAG), type 8 (SCA8, CTG / CAG), ty pe 10 (SCAIO, ATTCT / ATTTT), type 12 (SCA12, CAG / CTG), type 17 (SCA17, CAG / CAA), type 27B (FGF14, GAA), ty pe 31 (SCA31, TGGAA), ty pe 36 (SCA36, GGCCTG), type 37 (SCA37, ATTTC / ATTTT), and type 51 (SCA51); myotonic dystrophy type 1 (DMPK, CTG) and type 2 (CNBPIZ 9, CCTG); Fragile X syndrome (FMRI, CGG) with fragile X-associated tremor / ataxia syndrome (FXTAS, CGG premutation) and fragile X-associated primary- ovarian insufficiency- (FXPOI, CGG premutation); the hexanucleotide expansion disorders such as C9orf72-associated086013-0589445amyotrophic lateral sclerosis and frontotemporal dementia (C9orf72, GGGGCC); neuronal intranuclear inclusion disease (NOTCH2NLC. CGG); oculopharyngeal myopathy and leukoencephalopathy (OPML, CGG), oculopharyngodistal myopathy types 1-3 (0PDM1-3, CGG), and related CGG repeat phenotypes; cerebellar ataxia, neuropathy, and vestibular areflexia syndrome (CANVAS) (RFC1, AAGGG); the familial adult myoclonic epilepsies (e.g., FAME1, FAME2, FAME3, FAME4, FAME6, FAME7) caused by pentanucleotide repeats such as TTTCA, TTTTA, ATTTC, and ATTTT in intronic regions; Fuchs endothelial comeal dystrophy (TCF4, CTG / CAG intronic repeat); autosomal dominant tubulointerstitial kidney disease (MUC1, mono-C frameshift repeat); X-linked intellectual developmental disorders / Partington syndrome (ARX, GCG / CGC repeats); panhypopituitarism with growth hormone deficiency (SOX3, GCN); holoprosencephaly type 5 (ZIC2, GCN); VACTERL association (ZK'3. GCC); hereditary sensory and autonomic neuropathy type VIII (PRDM12. GCC); Fragile XE syndrome (AFF2, CCG) and other small-repeat disorders documented in loci such as PRNP or MARCHF6.

[0148] The terms “Huntington’s Disease” and “HD” refer to a genetic neurodegenerative disorder associated with an expanded CAG repeat in the huntingtin (HTT) gene, and characterized by progressive neurological and clinical changes. Huntington’s Disease is an example of a repeat expansion disorder. In the context of the present claims, Huntington’s Disease progression can be represented or evaluated using one or more simulated outcomes, including repeat length dynamics, cell survival, and / or predefined disease milestones.

[0149] The term “CAP- 100”, in the context of Huntington’s Disease, refers to a standardized CAG- Age Product (CAP) score that is scaled so that a CAP value of 100 corresponds to the expected age of diagnosis / onset.

[0150] The term “genetic neurodegenerative disorder” refers to a neurodegenerative disorder having a genetic component, including disorders associated with repeat expansions. Non-limiting examples include Huntington’s Disease and other repeat expansion disorders.

[0151] The term “disease progression” refers to advancement of a disease state over time, which can be evaluated using clinical measures, biomarkers, imaging measures, functional measures, and / or disease staging systems, and which can be approximated or represented in a simulation.086013-0589445

[0152] The term “natural disease progression"’ refers to disease progression in the absence of a therapeutic treatment or intervention being applied, including in a simulation configured to function as a control.

[0153] The term “control’" refers to a comparator, including a cell collection or simulation instance configured to simulate disease progression without therapeutic treatment, for purposes of comparing outputs to a treated population.

[0154] The term “treated” refers to a cell object, cell collection, or simulated instance in which a therapeutic is applied, administered, or modeled as administered.

[0155] The term “untreated” refers to a cell object, cell collection, or simulated instance in which a therapeutic is not applied, administered, or modeled as administered, and which can function as a control.

[0156] The term “cell survival” refers to a measure of whether a modeled cell remains viable / alive over simulated time, and / or a probability or quantity of modeled cells remaining viable / alive at a given time. In embodiments, cell survival is assessed by whether a cell object has not reached a defined death condition, threshold, or milestone in the simulation.

[0157] The term “healthy” when used in reference to a cell object, refers to a cell object satisfying one or more criteria indicative of a non-pathologic or less-pathologic simulated state, such as a repeat length remaining below a threshold at a specified time interval.

[0158] The term “threshold” refers to a predetermined value or set of values used to classify or categorize simulated outcomes, including classifying a cell object as healthy based on whether a repeat length remains below the threshold.

[0159] The term “disease milestone” refers to a predefined biological, clinical, imaging, biomarker, functional, or staging endpoint that is used to assess disease progression. In embodiments, a disease milestone includes onset of volumetric changes to the caudate and putamen, changes in cerebrospinal fluid biomarkers such as neurofilament light, onset or worsening of motor symptoms, and / or a specified disease stage.

[0160] The term “HD ISS” refers to the Huntington's Disease Integrated Staging System (HD-ISS) or another Huntington’s Disease staging system, including stages 1, 2, and / or 3, as used for characterizing disease progression.086013-0589445

[0161] The term “therapeutic effect'’ refers to one or more differences between (i) a simulation of a treated cell collection and (ii) a simulation of an untreated / control cell collection, including differences in somatic instability, repeat length dynamics, cell survival, and / or timing of one or more disease milestones.

[0162] The term “therapeutic benefit” refers to a therapeutic effect that is beneficial relative to a comparator (e.g., untreated / control), including a temporal delay in reaching a predefined disease milestone, increased cell survival, and / or reduced somatic repeat expansion.

[0163] The term “temporal delay” refers to a difference in simulated time at which a predefined disease milestone is reached (or is predicted to be reached) between (i) a treated population and (ii) an untreated / control population. In embodiments, a temporal delay is calculated as a treated time-to-milestone minus an untreated time-to-milestone, such that a positive value indicates that the treated population reaches the milestone later than the untreated / control population.Therapeutics, Transduction, and Treatment Status

[0164] The term “therapeutic” refers to an agent, composition, modality, intervention, or treatment intended to produce a beneficial effect with respect to a disease or biological process. In the context of the present claims, a therapeutic can be applied in a simulation by modifying one or more modeled parameters (e.g., reducing a somatic expansion rate) and / or by applying a modeled transduction rate to a subset of cell objects.

[0165] The term “therapeutic treatment” refers to applying a therapeutic. In embodiments, “therapeutic treatment” in a simulation includes updating one or more simulation parameters and / or states to represent administration, uptake, expression, activity, or other effect of a therapeutic.

[0166] The term “treatment status” refers to an indication or flag associated with a cell object representing whether the cell object is treated or untreated. In embodiments, treatment status includes a transduction status indicating whether a cell object has received, or is modeled as having received, a gene therapy or other therapeutic.

[0167] The term “cell transduction” refers to delivery of a nucleic acid and / or gene therapy payload into a cell, including by viral vector, non-viral vector, or other delivery’ modality,086013-0589445resulting in uptake and / or expression of the pay load. In the context of a simulation, cell transduction can be modeled as an event or state transition for a cell object.

[0168] The term “rate of cell transduction” refers to a parameter describing how cell transduction occurs across a population, including a proportion of cells transduced over a defined time period, a probability that a given cell object becomes transduced, and / or a distribution of transduction events over simulated time.

[0169] The term “transduction status” refers to a state or attribute of a cell object indicating whether the cell object has undergone a transduction event (e.g., received a gene therapy pay load) in the simulation.

[0170] The term “transduction event” refers to an event in which a cell becomes transduced with a gene therapy payload or vector. In embodiments, a modeled therapeutic transduction event refers to a simulated event that changes a cell object’s treatment status and / or transduction status according to a rate of cell transduction.

[0171] The term “gene therapy” refers to a therapeutic modality involving administration of a nucleic acid to a subject or cell to achieve a therapeutic effect, including by expression of a transgene, suppression of a target gene, editing of a genomic locus, and / or modulation of gene expression, and including delivery by viral or non-viral vectors. In embodiments, gene therapy includes a CRISPR-based method to excise excess CAG repeats, that would also be expected to affect SI.Gene Targets and Example Modalities

[0172] The term “target gene product” refers to a biological product associated with a target gene, including a transcript (e.g., mRNA), a protein, and / or an isoform, variant, fragment, or processed form thereof. References to reducing “expression” of a target gene product include reducing the abundance and / or activity of the target gene product, and can include reducing transcription, translation, stability, and / or functional activity relative to a comparator.

[0173] The terms “knockdowor”, “knock down". “KD”, "knockdow n value” refer to a reduction in expression of a target gene product relative to a comparator (e.g., relative to an untreated / control condition). Knockdown can be measured at the level of RNA (e.g., mRNA) and / or protein, and can be expressed as a percent reduction (e.g., 50% knockdown). In the086013-0589445context of a simulation, knockdown can be modeled as a reduction in an expression variable and / or as a modification of one or more modeled biological parameters influenced by the target gene product, such as a modeled somatic repeat expansion rate, mutation rate, and / or disease progression rate.

[0174] The term “knockdown parameter” refers to a model input variable representing a degree of knockdown of a target gene product in treated cell objects. In embodiments, a knockdown parameter is expressed as a fraction or percentage reduction relative to an untreated / control condition and is used to modify one or more simulated processes (e.g., to reduce a modeled CAG repeat expansion rate for treated cell objects).

[0175] The term “MSH3” refers to the gene and / or protein MutS homolog 3, including naturally occurring forms, variants, and homologs. In embodiments, reducing MSH3 expression refers to reducing transcription and / or translation of MSH3 relative to a comparator.

[0176] The term “siRNA” refers to a small interfering RNA that can reduce expression of a target gene via RNA interference. In embodiments, an siRNA targeting MSH3 refers to an siRNA having sequence complementarity sufficient to reduce MSH3 expression.

[0177] The term “miRNA” refers to a microRNA or microRNA-based construct that can reduce expression of a target gene, including by targeting a messenger RNA and / or by incorporation into an RNA-induced silencing complex. In embodiments, a miRNA targeting MSH3 refers to a miRNA having sequence complementarity sufficient to reduce MSH3 expression.

[0178] The term “antisense oligonucleotide” refers to an oligonucleotide that is complementary to at least a portion of a target nucleic acid and is capable of reducing expression of a target gene and / or modifying splicing of a target transcript. In embodiments, an antisense oligonucleotide targeting MSH3 refers to an oligonucleotide complementary to at least a portion of an MSH3 transcript.

[0179] The term “zinc finger protein” refers to a protein comprising one or more zinc finger domains capable of binding a target nucleic acid sequence and which can be engineered or used to modulate gene expression, including by recruitment of transcriptional repressors and / or epigenetic effectors.086013-0589445

[0180] The term “CRISPR-based means” refers to a CRISPR system or component (e.g., a Cas nuclease or nuclease-dead Cas (dCas) coupled with a guide RNA) configured to edit a genomic locus and / or modulate gene expression, including to reduce or silence expression of a target gene such as MSH3.

[0181] The term “epigenetic effector” refers to a protein, domain, or molecule capable of modifying chromatin and / or epigenetic marks (e.g., DNA methylation, histone modification) to modulate gene expression. In embodiments, an epigenetic effector is recruited to a target gene locus to reduce or silence gene expression.

[0182] The term “characteristics of the therapeutic” refers to one or more parameters of a therapeutic that can be modeled in a simulation, including dose, potency, efficacy, onset time, duration, transduction rate, expression level, tissue targeting, and / or a modeled effect on somatic expansion rate.Treated-Fraction, TPP, Candidate Configurations, and Decision

[0183] The terms “treated-fraction”, “treated fraction”, “treated-fraction value”, “treated-fraction parameter” refer to a proportion of cell objects in a cell collection that are designated as treated (e.g., as receiving simulated therapeutic treatment), optionally via a modeled rate of cell transduction. In embodiments, treated-fraction can be expressed as a percentage or as a ratio N_treated / N_total, where N_treated is a number of treated cell objects and N_total is a total number of cell objects in the population. A treated-fraction parameter refers to a model input variable representing the treated-fraction value and which can be varied (e.g., across a model parameter space).

[0184] The term “target product profile” and “TPP” refer to a set of characteristics, specifications, and / or performance criteria for a therapeutic, including minimum and / or optimal values for one or more parameters (e.g., treated-fraction, knockdown, transduction rate, dose, durability, safety, and / or predicted therapeutic benefit). A target product profile can be used to guide, screen, or qualify candidate therapeutics and / or candidate therapeutic configurations.

[0185] The term “minimum target product profile” refers to a target product profile comprising one or more minimum criteria and / or threshold values that are sufficient to achieve a defined outcome (e.g.. a predicted therapeutic benefit comprising a temporal delay greater than a predefined duration in reaching a predefined disease milestone relative to an086013-0589445untreated / control population). In embodiments, a minimum target product profile includes a minimum treated-fraction value and a minimum knockdown value determined based on executing simulations across a parameter space.

[0186] The term “candidate therapeutic configuration” refers to a specified set of characteristics, parameters, and / or settings describing (i) a therapeutic and / or (ii) application of the therapeutic, including in a simulation. A candidate therapeutic configuration can include, without limitation, a therapeutic modality, dose, route of administration, timing of administration, rate of cell transduction, treated-fraction value, knockdown value, target gene product, and / or other modeled therapeutic characteristics.

[0187] The term “evaluating” refers to assessing or determining, using one or more simulations, whether a candidate therapeutic configuration is predicted to achieve one or more criteria, including whether a predicted therapeutic benefit meets or exceeds a threshold (e.g.. a temporal delay greater than 5 years in reaching a predefined Huntington’s Disease milestone relative to an untreated / control population).

[0188] The term “selecting” refers to identifying, choosing, flagging, ranking, prioritizing, and / or otherwise designating one or more candidate therapeutic configurations for further development, a clinical application, and / or a clinical trial, optionally based on evaluation results and / or satisfaction of a target product profile.

[0189] The term “rejecting” refers to excluding, not selecting, and / or otherwise designating a candidate therapeutic configuration as failing to satisfy one or more criteria (e.g., failing to satisfy a minimum target product profile and / or failing to achieve a predicted therapeutic benefit threshold).

[0190] The term “modifying” refers to adjusting, changing, tuning, and / or updating one or more parameters associated with a therapeutic and / or candidate therapeutic configuration, including modifying one or more simulation inputs such as a rate of cell transduction, treated-fraction parameter, knockdown parameter, dose parameter, timing parameter, and / or other modeled therapeutic characteristic.

[0191] The term “minimum treated-fraction value” refers to a smallest or low est treated-fraction value, within a set of evaluated treated-fraction values, that satisfies a specified086013-0589445criterion or set of criteria, such as producing a predicted therapeutic benefit comprising a temporal delay greater than a predefined duration relative to an untreated / control population.

[0192] The term “minimum knockdown value” refers to a smallest or lowest knockdown value, within a set of evaluated knockdow n values, that satisfies a specified criterion or set of criteria, such as producing a predicted therapeutic benefit comprising a temporal delay greater than a predefined duration relative to an untreated / control population.Individuals and Clinical Measurement

[0193] The term “individual” refers to a human subject or patient. In embodiments, an individual is a candidate for treatment with a therapeutic, and / or is represented by a modeled cell collection.

[0194] The term “population of individuals” refers to two or more individuals, optionally defined by one or more criteria such as age, repeat length, genotype, phenotype, and / or eligibility' criteria for treatment or a clinical trial.

[0195] The term “treatable individual” and “treatable population” refers to an individual or population of individuals identified as having characteristics (e.g., age and repeat length) suggesting that treatment with a therapeutic would provide a predicted therapeutic effect and / or benefit, as determined using one or more simulations and / or selection criteria.

[0196] The term “clinical trial” refers to a prospective study in human subjects designed to evaluate one or more effects of a therapeutic, including safety7and / or efficacy.

[0197] The term “eligibility for treatment” refers to satisfaction of one or more criteria for receiving a therapeutic and / or enrolling in a clinical trial, including criteria based on age, repeat length, predicted therapeutic effect, cost, and / or other clinical or logistical factors.

[0198] The term “magnetic resonance imaging (MRI)” refers to an imaging modality that can be used to assess anatomical and / or functional features of the brain, including volumetric measurements of structures such as the caudate and putamen.

[0199] The term “cerebrospinal fluid (CSF)” refers to the fluid surrounding the brain and spinal cord, which can be sampled and analyzed for biomarkers of neurodegeneration.086013-0589445

[0200] The term “neurofilament light (NFL)” refers to neurofilament light chain protein (often abbreviated NfL or NFL), a biomarker that can be measured (e.g., in CSF) and associated with neuroaxonal injury and neurodegeneration.ExamplesEXAMPLE 1Model Framework and Results

[0201] Model Model inputs (FIGS.3A-3B) may include: 1. patient demographics: germline CAG number and age at the time of intervention; and 2. intervention characteristics: percent of transduced medium spiny neurons and level of MSH3 knockdown; among other inputs, as described herein. Model outputs may include: 1. prediction of treatment effects of a therapeutic; 2. identification of the optimal HD patient populations that are most likely to benefit from the therapeutic; 3. guidance on optimal target product profile for the therapeutic; and 4. adaption with availability of new natural history data; among other outputs, as described herein.

[0202] FIG. 9 pertains to simulating somatic instability in HD patient MSNs, individuals, and populations. FIG. 9, panel A, shows that each simulated medium spiny neuron (MSN) contains a defined germline CAG repeat length that can be modified by time and targeting status (targeted or not targeted). FIG. 9, panel B, represents an example simulation of a single MSN starting with a germline 40 CAG repeat (y-axis) tracked across time from ages 0-100 (x-axis). CAG repeat length is plotted at each simulated age on the y-axis. The right panel depicts the CAG repeat lengths of four simulated MSNs each with different SI rates over time. FIG. 9, panel C, pertains to the simultaneous simulation of MSNs across age models disease progression in MSNs of a simulated individual with HD. During simulation, cells are classified as transcriptionally healthy at <150 CAGs, dysregulated at >150 CAGs, or dead at >300 CAGs. FIG.9, panel D, shows a single frame of the simulation captured at age 40 from an animation of 3,000 MSNs across ages 0-100. The right panel depicts the percentage of healthy MSNs simulated across ages 0-100. The slope of the curve is reflective of the speed of MSN loss and rate of disease progression. FIG.9, panel E, pertains to simulation of the HD patient population and captures the spectrum of patients with a range of germline CAG repeat lengths and a range of ages. FIG.9, panel F, pertains to a simulation in which three patients with germline CAG lengths of 40, 45, and 50 are simulated and086013-0589445percentages of healthy MSNs are tracked across ages. Panels A, C, and E were created using BioRender.com.

[0203] FIG. 2 shows a simulation pipeline schematic. FIG. 2, panel A, pertains to object classes core to the simulation which include “cell'’ and “person.” Objects of class “cell” each contain attributes including a defined CAG length, a binary therapeutic status indicator, and a year counter. Cell objects can be modified by the indicated functions. Objects of class “person” contain a single attribute, a collection of simulated MSNs. By assigning all cells to a “person” object, a group of cells can be manipulated together across ages in a simulated lifetime. Objects of class “person” are modifiable by functions that apply mutation events to the CAG repeat at the prescribed rate. This rate may be modified by the simulated addition of an Si-targeted therapy in which cells are designated as targeted randomly in accordance with the provided rates of therapeutic efficacy (% of MSNs targeted and rate of SI lowering by % of MSH3 knockdown). FIG. 2, panel B, shows that after class definition, cell objects must be instantiated. The number of created “cells” is user defined as is the germline CAG repeat length assigned to these cells. A “person” object is then instantiated and assigned a list of cell objects. FIG. 2, panel C, shows that a lifetime is then simulated by modifying the CAG length each year in a user provided lifespan according to the Handsaker two-phase linear model of CAG repeat expansion. Simulation of therapeutic intervention can be integrated according to user provided rates of MSNs impacted and SI lowering. FIG. 2, panel D, shows that Si-targeting therapies can be assessed across a population with ranges of germline CAG repeat lengths (e.g. 37-60) and ages at time of treatment (e.g. 5-60 years old) via a nested loop to iterate over both variable ranges, testing all provided CAG repeat lengths and ages at treatment combinations. Output measurements including therapeutic benefit (years of delayed onset) to a variety of HD landmarks (onset of HD-ISS Stage I, II, III, motor symptom onset and others) may be calculated this way.

[0204] At the base level of the model, individual MSNs are simulated as independent objects with uniquely defined features, including germline CAG repeat length, age, and the potential to be therapeutically modified (FIG. 9, panel A, and FIG. 2, panel A). As simulated MSNs accumulate somatic CAG repeats over time, they are classified as transcriptionally healthy (<150 CAGs), transcriptionally dysregulated (>150 and <300 CAGs), or dead (>300 CAGs) (FIG. 9, panel B). MSNs modeled with identical starting parameters show expected stochastic variation, following distinct somatic expansion086013-0589445trajectories due to their rates of SI and reflecting the annual probability of variable DNA mismatch repair events, including expansion, contraction, or no change in CAG repeat length (FIG. 9, panel B) (2). To represent the collective caudate nucleus and putamen (striatum) in an HD patient, MSN objects are grouped together and assigned to an HD “person” object (FIG. 9, panel C, and FIG. 2, panels A and B). The simulated HD “person” is configured to age; and, with each simulated year, the impact of CAG repeat expansion on each of the assigned MSN objects tracked (FIG. 9, panel D, and FIG. 2, panel C). In the example simulation, the model replicates the stochastic variation in accumulated somatic CAG repeat expansions across MSNs, with some MSNs exhibiting >150 CAGs and the majority remaining in the healthy range at <150 CAGs (FIG. 9, panel D). When the number of MSNs classified as transcriptionally healthy and dysregulated are tracked in a simulated person's striatum over many decades, the model recapitulates the characteristic late onset and progressive nature of HD after symptom onset. Finally, simulation across a spectrum of CAG repeat lengths and ages explores variation in rates of disease progression at the HD patient population level (FIG.9, panel E). As an example simulation, where germline CAG repeat lengths are 40, 45, or 50, a shift occurs in the slope of the line, representing fewer healthy MSNs at younger ages with increasing CAG repeat lengths. Thus, the known inverse correlation between germline CAG repeat length and age of disease onset in HD is recapitulated (i.e., individuals with longer germline CAG repeats have earlier ages of onset) (FIG. 9, panel F)

[0205] FIGS. 3A-3B illustrate modeling MSN CAG length and cell survival over a lifetime. In this example, two thousand (2.000) MSNs are modeled over their lifetime. Still frames representing this modeling at 20-year intervals were extracted from an animation spanning years 1-100 of a simulated HD patient lifetime. FIG. 3A, shows simulated somatic expansion in individual MSNs in an untreated individual at 20 year intervals. Each point represents a single MSN and position along the X axis represents simulated CAG length. Cells to the left of the vertical line in each image marking the 150 CAG threshold represent healthy cells while cells to the right of this vertical line represent unhealthy cells progressing toward cell death defined here as >300 CAGs. In FIG. 3B, simulated cells (cell objects) are shown as in (A), except here the addition of a hypothetical therapeutic construct is simulated with conservative performance characteristics (50% transduction and 50% MSH3 knockdown). Transduced cells are circled.086013-0589445 EXAMPLE 2Tracking a simulated percentage of healthy MS s over a lifetime with treatment and measurement of therapeutic benefit (FIGS.4A-4B, 5A-5C)

[0206] FIGS. 4A-4B illustrate tracking a simulated percentage of healthy MSNs over a lifetime with treatment. FIG. 4A, illustrates a percentage of healthy cells (<150 CAGs) tracked and plotted across a simulated lifetime (0-100 years). Shaded background panels represent the indicated HD-ISS stage and dotted lines represent the indicated important milestones in HD symptom progression. In FIG. 4B, the untreated line (i.e., the line that is closer to the x axis in this graph) is carried forward as a comparator and a second "treated" line tracking the percentage of healthy cells in the treated cohort is added. The shaded background panels and dotted lines reflect the stages and HD landmarks associated with the line. The numbers at the bottom of both plots (along the x axis) indicate the age at which each HD-ISS stage is reached.

[0207] FIGS. 5A-5C illustrates measurement of therapeutic benefit as years of delay to HD landmark or HD-ISS stage. These figures illustrate how, using the model described herein, therapeutic benefit can be tracked by determining the age and healthy cell number coinciding with a HD landmark event or HD-ISS stage. Here, events and stages were assigned based on the literature and data from longitudinal studies of HD patient cohorts representing a variety of germline CAG lengths.EXAMPLE 3Simulations of therapeutic targeting of somatic instability predict delayed progression of HD and study design

[0208] To simulate the impact of an Si-targeted therapy, the rate of CAG repeat expansion in treated MSNs was modified by two therapeutic parameters: the percentage of targeted MSNs and the percentage of target reduction (MSH3 KD), using the reported proportional linear relationship between MSH3 levels and SI index (11, 1 ). As a demonstration, therapeutic intervention at age 32 was simulated with 50% MSH3 KD in 50% of MSNs (FIGS. 3A-3B, 4A-4B, and 5A-5D). These parameters were applied to a simulated individual with an inherited germline CAG repeat length of 40, and simulations conducted with and without treatment to model healthy and dysregulated MSNs from ages 20 to 80 years old (FIGS. 3A-3B, and 4A-4B). By comparing the percentage of healthy MSNs output from086013-0589445simulations without treatment (FIG.4A), and with treatment (FIG.4B), therapeutic benefit is predicted. Therapeutic benefit is defined as the difference in years between the age at which a measurable HD stage-specific landmark event intersects the percentage of healthy MSNs (grey dotted line in FIGS.5A-5C) in an untreated person and the age at which the therapeutically modified percentage of healthy MSNs (show n by the purple dotted line and purple arrow in FIGS.5A-5C) reaches that same value in a treated person. HD clinical landmarks and HD stages were assigned based on longitudinal studies of HD patient cohorts representing a variety of germline CAG lengths (19). Here, therapeutic benefit was calculated using the predicted age of onset for volumetric changes in caudate and putamen (FIG.5A), the onset of clinical symptoms (motor and cognitive) by HD-ISS stage II (FIG. 5B), and the approximate age of onset for motor symptoms as defined by CAP- 100 (FIG. 5B). Finally, therapeutic impact measured as change in biomarkers can be predicted through integration of natural history datasets from longitudinal studies of HD patients. By indexing biomarker readouts by CAG repeat length and age, biomarker scores are integrated with the model and predict change in biomarkers over time in simulations of treated and untreated HD patient populations. With treatment at age 32, the rate of cUHDRS (composite Unified Huntington Disease Rating Scale) decline is reduced in the treated condition compared to natural history (untreated), resulting in a steady separation of cUHDRS change over baseline scores across the available range of natural history data points (FIG. 5D) (20). In summary, simulating the effect of an Si-targeted therapy over a patient’s lifetime characterizes therapeutic benefit as years of delay in onset of several HD landmarks as well as reduced decline in cUHDRS.

[0209] FIGS. 3A-3B, 4A-4B, and 5A-5D pertain to simulating MSH3 knockdown on MSN survival over a patient lifetime. FIG. 3A-3B are still frames from an animation of untreated (FIG. 3A) and treated (FIG.3B) HD patients are shown from example simulations of 2000 MSNs that depict somatic CAG repeat expansion in individual MSNs at 20-year intervals from 1-100 years. (FIG. 3 A) Each point represents an untreated MSN and simulated CAG length (x-axis). The red line represents the 150 CAG threshold for transcriptionally healthy MSNs (<150 CAG) and dysregulated unhealthy MSNs (>150 CAG) progressing toward cell death (>300 CAGs). In FIG. 3B, simulated MSNs are shown with Si-targeted therapy administration at 32 years old and engaging 50% of MSNs with 50% MSH3 knockdow n. Targeted MSNs are circled in pink. In FIG.4A, the percentage of untreated healthy MSNs (<150 CAGs) are tracked and plotted across a simulated lifetime (0-100 years). Shaded background panels represent the indicated HD-ISS stage, age when a HD-ISS stage is086013-0589445reached, and dotted lines represent a HD milestone event (measurable volumetric change and CAP- 100 equivalent age). In this figure. Stage 1 begins at 44 years. Stage 2 begins at 60 years, and Stage 3 begins at 65 years. In FIG.4B, a treated line (upper line on the graph) is compared to the untreated line (lower line on the graph) from FIG.4A and tracks the percentage of healthy treated MSNs after administration at 32 years old. In this figure, Stage 1 begins at 48 years, Stage 2 begins at 69 years, and Stage 3 begins at 79 years. In FIGS.5A-5C The model tracks therapeutic benefit by determining the age and healthy MSN number coinciding with a HD landmark event (onset of striatum volume change, motor symptoms) or HD-ISS stage. FIG.5D shows that the model enables prediction of biomarker changes based on natural history data from HD cohorts. Here the predicted change in cUHDRS change from baseline between simulated treated and Enroll-HD natural history cohorts is evaluated (untreated).

[0210] The overall study design involved computationally simulating CAG repeat expansion in MSNs and predicting clinical benefit after targeting SI with a MSH3-lowering therapy. Preclinical proof-of-concept studies demonstrated AAV-mediated miMSH3 expression and lowering of MSH3 in NHPs and HD mice. AAV-mediated MSH3 lowering rates in NHP MSNs were used to predict clinical benefit across a broad range of HD patients. Biological MOA and impact of Msh3 lowering on somatic CAG repeat instability was evaluated in HD mice.EXAMPLE 4Treatable range of HD patients varies with transduction efficiency and MSH3 knockdown and targeting somatic instability early in HD is predicted to provide greater clinical benefit (FIG.6)

[0211] FIG.6 illustrates how a treatable range of HD patients varies with transduction efficiency and MSH3 knockdown. FIG.6 illustrates heatmaps depicting the simulated range of therapeutic benefit for patients treated with a therapeutic. Each heatmap represents a range of patients with germline CAG lengths from 37-60 CAGs along the Y axis and age at treatment represented along the X axis. The shading of each cell in the heatmap represents the degree of therapeutic benefit as measured by years of delay to reaching CAP- 100 equivalent. Shading in the upper right of each heatmap represents less than five years of benefit, the darkest shading indicates that CAP- 100 equivalent was never reached, and the gradient shading indicates a benefit between 5 and 50 years. FIG. 6, panel A, illustrates the range of086013-0589445therapeutic benefit achieved with a simulated therapeutic construct transducing only 25% of MSNs and achieving only 25% MSH3 knockdown per cell. FIG. 6. panel B, illustrates three heatmaps depicting the range of therapeutic benefit achievable with 50% MSN transduction and the indicated levels of MSH3 knockdown. FIG. 6, panel C, illustrates three heatmaps depicting the range of therapeutic benefit achievable with 50% MSH3 Knockdow n and the indicated variable levels of MSN transduction.

[0212] To probe the viability of an Si-targeted therapy, predicted therapeutic benefit was characterized across the HD patient population. Therapeutic benefit w as tracked across a range of germline CAG repeat lengths from 37 to 60 CAGs and ages at intervention from 5 to 60 years of age (FIG. 6, panels A-C). A minimum therapeutic benefit of 5 years was posited as a reasonable threshold for the performance of an Si-targeted therapy to show measurable efficacy in an HD patient clinical trial. To define the range of HD patients meeting this efficacy threshold, outcomes for an Si-targeted therapy based on therapeutic performance parameters, percentage of MSNs targeted and percentage of MSH3 KD were simulated. Initially, the impact of 25% MSN targeting with 25% MSH3 KD was tested. The model predicts five years of therapeutic benefit for HD patients with low germline CAG repeat lengths that undergo treatment early in life (FIG. 6, panel A). MSH3 KD levels of 50%, 75%, and 100% were then simulated and the percentage of targeted MSNs held at 50% (FIG.6, panel B), or, simulated variable MSN targeting rates of 50%, 75%, and 100%, and kept MSH3 KD levels at 50% (FIG. 6, panel C). The output reveals that neuron survival is maximized by early intervention, but also that the reduction of MSH3 levels in MSNs is more impactful than reducing modest amounts of MSH3 in all cells. Thus, MSNs targeted by an Si-targeting therapy will still experience rapid CAG repeat expansion if the level of MSH3 KD is minimal or if the targeted MSN already has an advanced CAG repeat length.Conversely, if SI is prevented in a low- number of MSNs through robust MSH3 KD, those MSNs could indefinitely persist in a healthy CAG repeat length. Collectively, these simulations bracket 50% MSN targeting with 50% MSH3 KD as targets for improved therapeutic impact for a large cross section of the HD patient population (as defined by a range of CAG lengths and ages).

[0213] FIG. 6 pertains to predicting therapeutic benefit across a spectrum of simulated HD patients. FIG. 6, panels A-C, are heatmaps depicting the simulated therapeutic benefit for a SI lowering therapy in a range of treated patients with germline CAG lengths from 37-60086013-0589445CAGs (y-axis) and ages at treatment from 5-60 years of age (x-axis). The degrees of predicted therapeutic benefit as measured by years of delay to CAP-100 equivalent are depicted in the heatmaps (shown in the heatmap key).EXAMPLE 5Artificial miRNAs potently reduce MSH3 expression in cells and wildtype mice

[0214] Artificial miRNAs were designed to target both NHP and human MSH3 mRNA (miMSH3). Two candidates, miMSH3-06 and miMSH3-l 1, emerged from the in vitro screening for potency and strand bias, showing minimal passenger strand loading and potent reduction of endogenous human and mouse MSH3 mRNA in HEK293 cells and N2a cells, respectively (FIGS. 14A-14E) (23).

[0215] FIGS. 14A-14E pertain to in vitro screening identifies potent artificial miRNAs targeting MSH3. FIG. 14A is a schematic depicting MSH3 miRNA (miMSH3) in silico design with siSPOTR (22) and in vitro miMSH3 candidate screening workflow. FIG. 14B shows that twenty -nine miMSH3 guide strand sequences embedded in the miR-30 scaffold were screened by dual-luciferase knockdown assay in HEK293 cells after co-plasmid transfection of Renilla luciferase / firefly luciferase and miMSH3 expression plasmids. The miHDSl (HTT-targeting miRNA) guide strand target was included as a knockdown positive control for the assay. Renilla luciferase (guide stand target plasmid) expression was normalized to firefly luciferase expression. FIG. 14C shows that six potent miMSH3s and three low potent miMSH3s (miMSH3-12, miMSH3-15. and miMSH3-21) were screened for passenger strand knockdown by dual-luciferase assay. FIG. 14D shows that nine miMSH3 sequences were screened for knockdown of endogenous human MSH3 mRNA normalized to GAPDH in HEK293 cells by RT-qPCR. FIG. 14E show s that six miMSH3 sequences screened for knockdown of endogenous mouse Msh3 mRNA normalized to ActB in mouse N2a cells by RT-qPCR. N=3 independent experiments with three technical replicates, data are shown as the mean, and error bars represent standard deviation (SD). FIG. 14A was created using BioRender.com.

[0216] For in vivo studies, miRNAs were packaged in the AAV capsid variant, AAV-DB-3 (18). The miMSH3-06 and miMSH3-ll guide sequences were tested in the miR-30 scaffold (AAV-DB-3. miR30-miMSH3-06, AAV-DB-3.miR30-miMSH3-ll) and miR-451 scaffold (AAV-DB-3.miR451-miMSH3-06, AAV-DB-3.miR451-miMSH3-l 1). miMSH3-06 and miMSH3-l 1 embedded in the miR-30 scaffold outperformed the same miMSH3s embedded086013-0589445in the miR-451 scaffold, demonstrating greater Msh3 mRNA KD in the striatum of WT mice after intrastriatal delivery’ (FIGS. 15A-15C). Finally, miMSH3-06 and miMSH3-ll demonstrated efficient miRNA processing with high 5 ’-end accuracy (>90%) and favorable guide-to-passenger strand ratios in HEK293 cells (>9:1) and WT mouse striatal tissues (>85:1) (FIGS. 16A-16B).

[0217] FIGS. 15A-15C, miR-30 and miR-451 as scaffolds for miMSH3 mediated MSH3 reduction. FIG. 15A shows human MSH3 mRNA reduction with miRNA miR-30 scaffold relative to no miRNA control in HEK293 cells by RT-qPCR. miR-30.miHDSl targeting HTT was included as positive control. FIG. 15B shows human MSH3 mRNA levels after miRNAs with miR-451 scaffold. miHD12 targeting HTT was included as positive control. For FIG.15A and FIG. 15B, GFP-positive cells were sorted by FACS prior to RT-qPCR. N=3 independent experiments with three technical replicates and error bars represent SD. FIG. 15C shows that C57BL / 6J wildtype mice were bilaterally infused in the striatum at 8 weeks old with AAV-DB-3-miR-30.miMSH3 or AAV-DB-3. miR-45 l.miMSH3 at 5E10 vg or 1E11 vg with co-delivery of AAV-DB3-mTFPl at IE 10 vg to guide microdissection of transduced tissue 4 weeks later. Mouse Msh3 mRNA levels were measured by RT-qPCR and were normalized to ActB and plotted relative to vehicle control. N=3 mice per group shown as mean and error bars represent SD. FIG. 15C was created using BioRender.com.

[0218] FIGS. 16A-16B pertain to favorable miRNA processing in HEK293 cells, mouse striatum, and NHP putamen. FIG. 16A shows the percent miR-30 mature guide strands with accurate 5 '-end processing in the seed sequence, P2-P8. following expression in HEK293 cells, mouse striatum and NHP putamen as assessed by small RNA sequencing. FIG. 16B shows the guide to passenger strand ratios miMSH3-06 and miMSH3-l 1 in HEK293 cells, mouse striatum and NHP putamen, using small RNA sequencing. N=3 independent experiments in HEK293 cells, N=3 WT mice, N=2 NHPs. Data are the mean and error bars represent SD.

[0219] As miRNA sequences can be partially complementary to unintended mRNAs, resulting in off-target gene expression effects, the off-target profiles of miMSH3-06 and miMSH3-l 1 were evaluated. Both miMSH3-06 and miMSH3-l 1 showed low off-target KD activity and robust KD o MSH3 mRNA in HEK293 cells (FIG. 17). miMSH3-06 showed fewer off-target hits by low fold change (<1 log2 fold change) (FIG. 17). Overall, these086013-0589445results supported the selection of miMSH3-06, hereafter called miMSH3, as a lead miRNA candidate for further study.

[0220] FIG. 17 pertains to low detectable off-target gene expression in HEK293 cells transfected with miMSH3-06. Volcano plots depict off-target differential gene expression determined by RNA sequencing of miMSH3-06 and miMSH3-ll transfected HEK293 cells, relative to the CMV-GFP control transfection. Data were cleaned using miHDSl control to remove shared transcriptional changes from plasmid transfection. Volcano plots indicate significance of observed differential gene expression on the x-axis (Log2 fold change) and amplitude of upregulated (right of zero (dotted line) on the x-axis) and down regulated (left of zero (dotted line) on the x-axis) gene expression on the y-axis (-log 10 P value). Dotted lines indicate zero on the x-axis, and dash-dotted lines indicate the thresholds for significance. Arrows indicate MSH3.EXAMPLE 6Viral vector AAV-DB-3.miMSH3 reduces MSH3 mRNA and protein levels in NHP caudate and putamen

[0221] A four-week in-life dose range finding (DRF) study was performed in rhesus macaques to identify effective doses of AAV-DB-3.miMSH3 for target reduction after intraparenchymal (IPa) globus pallidus (GP) infusion (FIG. 10A; Table 1). Bulk tissue analyses showed dose-dependent increases in AAV-DB-3 biodistnbution (FIG. 10B) and miMSH3 expression (FIG. 10C) in the GP, caudate, and putamen. AAV-DB-3 biodistribution was evident throughout brain regions, and there was limited detection in peripheral tissues; there were vector genomes (vgs) detected in the spleen and cervical lymph node at the mid and high doses, albeit at low levels (FIG. 18). In bulk GP tissue. MSH3 mRNA levels were reduced by 31%, 52%, and 47% relative to vehicle control for the low-, mid-, and high-dose groups, respectively (FIG. 10D). These reductions in mRNA corresponded to 22%, 47%, and 34% reductions in MSH3 protein levels, respectively (FIG.10E). In the caudate. MSH3 mRNA levels were reduced by 14%, 16%, and 39%. and MSH3 protein levels were reduced by 12%, 23%, and 28% at the low, mid, and high doses, respectively (FIGS. 10D-10E). Finally, in the putamen, MSH3 mRNA levels were reduced by 9%, 1 %, and 24% at the low', mid, and high doses, respectively. MSH3 protein levels were unchanged at the low dose and reduced by 19% at both the mid and high doses (FIGS.10D-10E). Together, MSH3 mRNA and protein were generally reduced in a dose-dependent manner in the caudate and putamen. In FIGS. 10B-10E, for each brain regions and from left086013-0589445to right, the tested doses are, per hemisphere (“hem"’), “Vehicle”, “7.5el0 vg / hem”, “4.1 ell vg / hem”. and “1.6el2 vg / hem”. Additionally, miMSH3 processing was favorable in the putamen of NHPs with >99% 5 ’-end accuracy and >30: 1 guide-to-passenger strand ratio in the NHP putamen (FIGS. 16A-16B).Table 1: Summary of Study Design for NHP (Rhesus Macaque) DRF Study

[0222] FIGS. 10A-10E pertain to target engagement and MSH3 lowering in NHP brain. FIG. 10A shows a study design for the nonhuman primate dose range finding study in rhesus macaques. FIG. 10B shows AAV Biodistribution (vector genomes per diploid genome) in the globus pallidus (GP), caudate nucleus, and putamen of vehicle and treated rhesus macaques by droplet digital PCR (ddPCR) after bilateral GP administration of AAV-DB-3.miMSH3-06.FIG. 10C shows miMSH3 expression levels (copies per pg of RNA) by stem-loop RT-qPCR.FIG. 10D shows MSH3 mRNA levels by RT-qPCR. FIG. 10E shows MSH3 protein levels086013-0589445by Jess capillary-electrophoresis immunoassay. Data are shown as group mean (N=2) with data points from each animal. FIG. 10A was created using BioRender.com

[0223] FIGS. 18A-18B pertain to high AAV biodistribution in NHP brain with minimal detection in other tissues. AAV genome biodistribution data by qPCR are show n as copies per microgram of DNA (left y-axis) and calculated AAV vector genomes (vgs) per diploid host genome (dg) (right y-axis) across brain tissues (FIG. 18A) and other tissues (FIG. 18B).Dose groups were 1.5ell vg, 8.2ell vg, and 3.2el2 vg shown from left to right, respectively, for each tissue. Biodistribution data not shown were below the limit of quantification (BLOQ). Data are shown as group mean (N=2) with data points from each animal. Samples from vehicle-treated control NHPs were below the limit of detection (BLOD). In plots, group means were calculated using a value of 0 for any result below- the limit of quantification (BLOQ).

[0224] Bulk tissue sampling confirmed biological activity of AAV-DB-3.miMSH3 and target engagement but likely underestimates MSH3 KD in MSNs as non-transduced cells (e.g. astrocytes, oligodendrocytes, etc.) dilute the MSN-specific KD effect from AAV-DB-3 tropism. Indeed, MSNs, while accounting for 85% of neurons in the striatum, are only 45% of the overall cell collection (24).

[0225] To quantify MSH3 mRNA levels specifically in DARPP32+ MSNs in the caudate and putamen of NHPs, single-cell RNAscope fluorescence in situ hybridization (FISH) was performed. An automated counting algorithm was used to quantify the puncta in the cytoplasmic compartment, affording quantification of MSH3 target engagement and KD. The cytoplasmic signal was prioritized because miRNAs primarily target mRNAs in the cytoplasm (25). Quantitative analysis of confocal images from RNAscope experiments showed dose-dependent decreases in cytoplasmic MSH3 mRNA in all MSNs in the striatum treated with AAV-DB-3.miMSH3 (FIG. 11). Compared to vehicle-treated animals, cytoplasmic MSH3 mRNA levels in striatal MSNs were reduced by 48%, 61%, and 94% in the low-, mid- and high-dose groups, respectively (FIG. 11). In this figure, the tested doses are. from left to right, “Vehicle’', “1.5el 1 vg”, “8.2el 1 vg”, and “3.2el 1 vg”. Lowering of MSH3 mRNA was also observed in MSN nuclei, suggesting a fraction of nuclear MSH3 mRNA was accessible to RNA interference (RNAi) KD or that KD occurred indirectly in the nuclear compartment, whereby both compartments reached an equilibrium (FIGS. 19A-19C).Together, these results demonstrate robust reduction of MSH3 mRNA in NHP striatal086013-0589445neurons, with the >50% KD observed in all MSNs by RNAscope FISH exceeding the predicted therapeutic threshold for successful translation of an Si-targeted therapy.

[0226] FIG. 11 pertains to dose-dependent lowering of MSH3 mRNA in striatal MSNs. FIG. 11 shows cytoplasmic MSH3 mRNA puncta count in DARPP32+ MSNs of the striatum (caudate and putamen) were determined with an automated cell segmentation and puncta counting algorithm. The percentages of cytoplasmic MSH3 mRNA knockdown (KD) in all MSNs of the caudate and putamen relative to vehicle control are shown above each treatment group. Total cell counts are displayed below each group. Data are shown as group mean (N=2) with data points from each animal and doses are shown as total vector genome (vg).

[0227] FIGS. 19A-19C, MSH3 mRNA lowenng in striatal MSNs. FIG. 19A shows RNAscope FISH quantitation of cytoplasmic MSH3 mRNA puncta counts in all MSNs of the striatum were determined with an automated cell segmentation and puncta counting algorithm. FIG. 19B shows RNAscope quantitation of nuclear MSH3 mRNA puncta counts in all MSNs of the striatum. FIG. 19C shows RNAscope quantitation of whole cell (cytoplasmic and nuclear) MSH3 mRNA puncta counts in all MSNs of striatum determined from the combined cell counts. Relative to vehicle control animals, the percentages of MSH3 mRNA KD in all MSNs are shown above each treatment group. Data are shown as group mean (N=2) with data points from each animal.EXAMPLE 7MSH3 knockdown levels in NHPs are predicted to provide clinical benefit in HD

[0228] Next, the observed RNAscope measurements of the in vivo therapeutic performance ° / oMSH3 KD quantified across all MSNs) in NHPs was used to benchmark the potential clinical efficacy of the low, mid. and high dose equivalents of our Si-targeted AAV gene therapy in our clinical simulation model. The MSH3 KD level in MSNs was set for each dose tested in NHPs across a range of inherited CAG repeat lengths (40-50) and ages at intervention (20-60 years old) (FIG. 12, panels A-D). This range captures most germline CAG repeat lengths present in the HD patient population (shown by the gradient CAG repeat length incidence indicator) (21). An arbitrary minimum threshold of 5 years of therapeutic benefit was applied, an efficacy target likely detectable in a clinical trial, to facilitate comparison of the NHP doses tested. The range of patients meeting the 5-year therapeutic threshold minimum are denoted as black outlined rectangles; for reference, they are shown on the low, mid. and high dose heatmaps (FIG. 12, panels A-D). Simulation of the MSH3 KD086013-0589445levels of 48.1%, 60.6%, and 94.0% in all MSNs at the low, mid, and high doses, respectively, demonstrated that all three KD levels and doses exceeded the minimum therapeutic threshold, indicating that each clinical dose equivalent is predicted to provide meaningful therapeutic benefit upon clinical translation (FIG. 12, panels B-D). Remarkably, all doses showed the potential to prevent a subset of patients from ever developing motor symptoms (>50 years of delay in onset to CAP-100 equivalent score; FIG. 12, panels B-D, rectangles). The predicted therapeutic effect of MSH3 KD was further demonstrated by substantial reduction in cUHDRS change over baseline by 53%, 40%, and 126% relative to natural history in the low, mid, and high doses, respectively, at three years post-treatment (FIG. 12, panel E).

[0229] FIG. 12 pertains to clinical therapeutic benefit predicted at all doses evaluated in NHPs FIG. 12, panels A-D, show heatmaps depicting the simulated range of therapeutic benefit for patients treated with an SI lowering AAV gene therapy over a lifetime of 100 years. Each heatmap represents a range of patients with germline CAG lengths from 40-50 CAG repeats (y-axis) and age at treatment (x-axis). The degrees of predicted therapeutic benefit as measured by years of delay to CAP-100 equivalent are depicted in the heatmaps.FIG. 12, panel A, shows the range of therapeutic benefit achieved after simulating 50% of MSNs transduced and 50% MSH3 KD per transduced MSN. This threshold is denoted by black bordered heatmap cells in all four heatmaps. FIG. 12, panels B-D, show Heatmaps depict the predicted range of therapeutic benefit for HD patients treated with the NHP DRF dose equivalents of AAV-DB-3.miMSH3 at the low dose (1.5el 1 vg) (FIG. 12, panel B), mid dose (8.2el 1 vg) (FIG. 12, panel C), and high dose (3.2el2 vg) (FIG. 12, panel D) based on the RNAscope FISH MSH3 mRNA KD levels detected in the NHP caudate and putamen (FIG. 11). CAG repeat length incidence shown by the gradient column (40). FIG.12, panel E shows that to characterize predicted therapeutic performance using cUHDRS across the observed percentages of MSH3 knockdown, these therapeutic performance levels were simulated in a hypothetical patient with 42 CAGs treated at the age of 46.EXAMPLE 8Viral Vector AAV-DB-3.miMSH3 reduces somatic instability (SI)

[0230] The effects of AAV-DB-3 -mediated delivery of miMSH3 on SI w ere next assessed in the HdhQl 11 KI mouse model of HD (FIG. 13A). The HdhQl 11 KI mouse exhibits robust somatic CAG expansion in the striatum, similar to observations in post-mortem human HD brains (1, 26). Study groups included untreated and vehicle-treated heterozygous086013-0589445HdhQl 11 KI mice, as well as heterozygous HdhQl 11 KI mice dosed with AAV-DB-3.miMSH3 at 5*109. 1.5*10105xlO10or 1.5xlOnvg per mouse via bilateral intrastriatal injection. Mice were injected at 8 weeks of age and euthanized 16 weeks later to assess biodistribution, target KD and SI.

[0231] FIGS. 13A-13D pertain to dose-dependent slowing of somatic CAG repeat instability in HdhQl 11 mice. FIG. 13A shows a study design for the pharmacology study in heterozygous HdhQl 11 KI mice. FIG. 13B shows Msh3 mRNA levels in the striatum of untreated and treated HdhQl 11 mice by RT-qPCR. Msh3 mRNA was normalized to ActB mRNA and shown as the group mean relative to vehicle control (N=7, 8 week untreated; N=7. 24 week untreated; N=9. vehicle; N=I0, 5e9 vg; N=10, 1.5el0 vg; N=I0, 5el0 vg; N=8, 1.5el 1 vg; one-way ANOVA with Dunnett’s post hoc analysis, **p < 0.01, ***p < 0.001, ****p < 0.0001). FIG. 13C shows fragment analysis traces depict the intensity of CAG repeat length alleles in untreated, vehicle and AAV-DB-3.miMSH3-06 treated mice by PCR and capillary electrophoresis. FIG. 13D shows somatic instability index in untreated and treated mice (N=10 / group; one-way ANOVA with Dunnett’s post hoc analysis, ****p < 0.0001). Data are shown as the mean and error bars represent SD. The percent reduction in SI relative to vehicle control is shown above each treatment group. FIG. 13A was created using BioRender.com.

[0232] In bulk striatal tissue, AAV-DB-3 biodistribution was dose-dependent and highly specific to the brain with minimal detection in peripheral tissues (FIG. 20). In this figure, for each tissue, the doses are, from left to right: “Vehicle”, “5e9 vg”, “1.5el0 vg”, “5el0 vg”, and “ 1.5e 11 vg”. Striatal Msh3 mRNA levels were reduced by 13%, 27%, 25%, and 23% at the 5xl09, 1.5xlO10, 5xl010and 1.5X1011vg / mouse doses, respectively, compared to 24-week-old vehicle-treated controls (FIG. 13B). Furthermore, striatal Msh3 mRNA levels in the 1.5xlO10, 5xlO10and ItyxlO11vg / mouse dose groups were significantly reduced compared to the 24-week-old vehicle-treated controls (FIG. 13B). These data suggest that increasing levels of striatal biodistribution do not correspond to a clear dose-response in Msh3 lowering, which appeared to be saturated above the 1.5xlO10vg / mouse dose level.

[0233] FIG. 20 pertains to dose-dependent AAV-DB-3 biodistribution in HdhQl 11 mouse brain tissues with minimal detection in peripheral tissues. This figure shows AAV biodistribution data by qPCR are shown as calculated AAV vector genomes (vgs) per diploid host genome (dg) (y-axis) across brain tissues and other tissues. Biodistribution not shown086013-0589445were below the limit of quantification (BLOQ). Data are shown as group mean (N=7, vehicle group; N=7, 5e9 vg group; N=7. 1.5el0 vg group; N=7. 5el0 vg; N=7. 1.5e 11 vg group) with data points from each animal. In plots, group means were calculated using a value of 0 for any result below the limit of quantification (BLOQ).

[0234] To assess the impact of AAV-DB-3.miMSH3. SI was measured. SI significantly increased from 8 to 24 weeks of age in untreated control striata and vehicle-treated 24-week-old controls did not differ significantly from 24-week-old untreated controls (FIGS. 13C-13D). AAV-DB-3.miMSH3 reduced the SI index in the striatum by 8%, 15%, 34%, and 46% at doses of 5xl09, 1.5*1010, 5xlO10and I.5xl0nvg / mouse, respectively, compared to vehicle-treated controls (FIG. 7C-D). Both the 5xlO10and 1.5x10“ vg / mouse dose levels significantly reduced SI relative to vehicle controls (FIG. 13D). In an independent validation study using miMSH3-l 1, intrastriatal delivery' of AAV-DB-3.miMSH3-ll in HdhQlll mice at a single dose of 5xlO10vg / mouse resulted in highly specific biodistribution and miMSH3 expression in the striatum and cortex. Tissue analyses showed significant lowering of striatal Msh3 mRNA by 20% and a concomitant significant reduction in striatal SI index by 43% (FIGS. 21A-21E). These studies show that AAV-DB-3.miMSH3 reduces Msh3 in HdhQl 11 KI mice supporting target engagement and effective reduction of SI.

[0235] FIGS. 21A-21E pertain to AAV-DB-3.miMSH3-ll reduces somatic instability in HdhQl 11 mice. FIG. 21A shows a study design for evaluating AAV-DB-3.miMSH3-ll in heterozy gous HdhQl 11 KI mice. Co-injection of AAV-DB-3.mTFPl was performed to enable microdissection of transduced mTFPl -positive striatal tissue. FIG. 21B shows AAV biodistribution by ddPCR in untreated and treated HdhQl 11 mice at 24 weeks old (N=3. untreated; N=6, AAV-DB-3.miMSH3-l 1 treated). FIG. 21C shows miMSH3-ll expression levels as measured by stem-loop RT-qPCR in untreated and treated HdhQl 11 mice at 24 weeks old (N=6, untreated; N=6 AAV-DB-3.miMSH3-ll treated). In FIGS. 21B-21C, for each tissue and from left to right, the data is shown for untreated mice and mice treated with AAV-DB-3.miSMH3-ll). FIG. 21D shows Msh3 mRNA levels in the striatum in untreated and treated HdhQl 11 mice at 24 weeks old by RT-qPCR. Msh3 mRNA was normalized to ActB and plotted relative to untreated controls (N=10 untreated, N=6 treated; unpaired t-test, **p < 0.01). FIG. 21E shows SI index in untreated and treated mice by PCR-based fragment analysis (N=10 untreated 8-week-old and 24-week-old mice; N=6, AAV-DB-3.miMSH3-ll treated 24-week-old mice; one-way ANOVA with Dunnett’s post hoc analysis, ****p <086013-05894450.0001). Data are shown as the mean and error bars represent SD. FIG.21A was created using BioRender.com.EXAMPLE 9Model outputs support therapeutic development

[0236] Simulation of CAG repeat expansion can be leveraged to generate outputs that support therapeutic development. These outputs include prediction of therapeutic benefit across the spectrum of HD patients, prediction of time to a threshold of detectable change in a clinical biomarker, and integration of these two metrics to define the ideal patient population for a clinical trial, which is characterized by both short time to detection and robust therapeutic benefit.

[0237] A core function of the model is to calculate therapeutic benefit as years of delay to the onset of a HD landmark event. FIG. 22A depicts the level of therapeutic benefit (Years of delay to the predicted onset of motor symptoms, CAP- 100 equivalent score) as a heatmap. Therapeutic benefit is individually simulated for each rectangle in the heatmap, with each rectangle corresponding to patient defining criteria. These criteria are patient age (on the x axis) and germline CAG repeat length (on the y axis). The simulations performed here utilized the minimum target product profile (50% MSH3 knockdown in 50% medium spiny neurons). Outputs in this format enable end users to assess the patient population predicted to achieve therapeutic benefit greater than a minimum therapeutic threshold. Here, a threshold of 5 years of therapeutic benefit is displayed. FIG. 22B pertains to time to a threshold of detectable change. Huntington’s disease biomarker data was integrated for cUHDRS and volumetric MRI from longitudinal natural history studies of HD patients. Integration of these biomarkers enables the assessment of the time in years for a threshold of volumetric change to be distinguishable between simulated therapeutically treated individuals and natural history data. Finally, the integration of results from both the years of benefit and time to detectable change was possible and allowed to define the optimal clinical trial population that is characterized by both short time to detectable clinical benefit and robust long therapeutic benefit potential, as shown in FIG. 22C. In this plot, Xs mark the range of patients meeting a threshold of less than 2 years to 1% change in vMRI and >5 years of therapeutic benefit (delay to the onset of CAP-100 equivalent) when treated with a therapeutic yielding 93% knockdown of MSH3 in all MSNs.086013-0589445EXAMPLE 10Discussion

[0238] As the HD therapeutic landscape evolves toward developing DNA repairmodulating strategies, there is a need for translational frameworks that quantitatively connect target engagement and impact on SI to predicted clinical outcomes. The model addresses this gap by integrating computational modeling with patient natural history and preclinical data to model predicted clinical benefit of targeting SI with a MSH3-lowering intervention. thereby predicting therapeutic thresholds and patient populations most likely to benefit.

[0239] The computational model includes the ability to assess therapeutic impact (e g., benefit). This is accomplished by simulating the lifetime trajectory of CAG lengths and associated viability of MSNs in "individuals’ in the presence and absence of SI modifying therapies. The model was applied across a broad HD patient population with variable germline CAG lengths and ages (2). By varying MSH3 KD levels, MSN coverage, treatment age, and inherited CAG repeat length, the model predicts that partial suppression of SI preserves healthy neurons. This outcome is validated by human GWAS data and mouse studies showing benefit from loss of MSH3 function. Notably, the model is broadly applicable to any Si-lowering approach, independent of modality, because the model defines a quantitative relationship between the degree of SI suppression in targeted MSNs and predicted disease modification. Thus, the framework establishes a scalable and modular platform for evaluating Si-targeted therapies, enabling investigators to test candidate interventions (antisense oligonucleotides, RNAi, CRISPR-based editing, proteins, small molecules, etc.). Importantly, even modest therapeutic performance (50% MSN targeting and 50% MSH3 knockdown) was predicted to delay motor onset by > 5 years for a w ide range of HD patients, demonstrating that a measurable and meaningful benefit could be achieved without complete target suppression. Finally, the model simulations of targeting SI, as well as data from human HD patients and animal models, indicates that lowenng MSH3 early in the course of disease would have the most impact on modifying disease before the CAG repeat expands beyond a pathogenic threshold (2, 4, 27). The model also suggests that treating later still provides meaningful benefit, albeit to a lesser degree.

[0240] The studies herein, inter alia, integrates the MSN-tropic properties of an engineered the AAV-DB-3 capsid variant for efficient and selective transduction of MSNs within the basal ganglia and deep layer cortical neurons (18), with a genetically validated Si-modifying086013-0589445target, MSH3, and a predictive modeling framework to guide preclinical development. The computational model identified the levels of MSH3 reduction required to yield clinical benefit, which was achieved in all doses evaluated in NHPs. Complimentary HdhQl 11 mouse studies demonstrated dose-dependent slowing of SI consistent with the predicted mechanism of action (MO A). Together, the convergence of empirical and simulated data illustrates how predictive translational modeling and preclinical studies can inform a therapeutic approach, enabling the establishment of efficacy thresholds, dose selection, and patient population criteria.

[0241] Si-targeting interventions such as MSH3 KD strategies could expand the therapeutic options beyond symptomatic HD patients to include presymptomatic HTT mutation carriers. In this population, slowing or stopping SI before the onset of Si-driven transcriptional dysregulation could preserve neuronal integrity and prevent disease.

[0242] Moreover, early intervention according to the model-informed thresholds could enable preventative gene therapy paradigms. Because SI occurs most rapidly (within decades of a patient’s lifetime) on the mutant allele, MSH3 lowering provides an indirect allele-selective strategy that could limit the expression of the toxic, hyperexpanded mHTT species, including HTTla, and formation of aggregates (28, 29). Additionally, SI targeting may lessen the need for HTT-lowering approaches, though combined strategies may yield greater benefit. Dual -targeting of MSH3 and mHTT, or additional targets such as PIAS1, could address complementary and independent mechanisms of neuronal toxicity (29, 30). Beyond MSH3, the model provides a generalizable framework for evaluating other Si-modulating targets (e.g., PMS1, FAN!) and combinatorial strategies targeting other DNA repair pathways (17, 31, 32). This modularity allows for expansion to assess additive or synergistic benefits from multimodal Si-targeting therapies.

[0243] More broadly, the computational and translational modeling framework can be adapted to other repeat expansion disorders where SI contributes to disease progression. MSH3 was identified as a modifier of X-linked dystonia-parkinsonism (XDP) onset, likely driven by SI of the CCCTCT repeat within the TAF1 gene (33, 34). Interestingly, XDP and HD share neuropathological features, primarily the loss of MSNs in the striatum. The similarities with HD suggest AAV-DB-3.miMSH3 as potentially beneficial for XDP. MSH3 was also identified as a modifier of SI of the CTG repeat in the DMPK gene and disease severity in myotonic dystrophy type 1 (DM1) (35, 36).086013-0589445

[0244] In summary integration of quantitative SI modeling with preclinical validation establishes a means for developing and predicting mechanism-based disease-modifying therapies for HD and other repeat expansion diseases.

[0245] Other repeat expansion diseases and disorders can be similarly modeled, and progression in the absence of or in the presence of a therapeutic treatment evaluated. For example, for a CAG / poly glutamine repeat expansion disease, a CAG repeat expansion rate can be defined for a given disease or disorder in a relevant cell or a plurality or population of relevant cells (i.e., cell objects). CAG / polyglutamine repeat expansion disorders include, for example, SCA1, SCA2, SCA3, SCA6. SCA7, SCA17, and dentatorubral-pallidoluysian atrophy (DRPLA). CAG expansion rates for cell objects receiving an effective therapeutic treatment can be modified to reflect reduced CAG repeat expansion. Other repeat expansion disorders involve non-CAG motifs, including, for example, CTG repeat expansions in myotonic dystrophy type 1 (DM1), CCTG repeat expansions in myotonic dystrophy type 2 (DM2), GAA repeat expansions in Friedreich ataxia (FRDA). CGG premutation expansions in fragile X-associated tremor / ataxia syndrome (FXTAS), and GGGGCC hexanucleotide repeat expansions in C9orf72-associated Amyotrophic lateral sclerosis (ALS) / Frontotemporal dementia (FTD). X-linked dystonia-parkinsonism (XDP) is associated with a polymorphic CCCTCT repeat within an SINE-VNTR-Alu (SV A) insertion in the TAF1 gene. For each such disorder, a repeat expansion rate corresponding to the relevant repeat motif can be defined for the relevant cell objects and modified to reflect the presence of a therapeutic intervention. Somatic repeat instability may vary depending on the repeat motif, genomic locus, repeat length, tissue and / or cell type. The model can parameterize such variability and simulate the impact of candidate therapies that reduce somatic expansion. The therapeutic effect on somatic instability based on the simulation thereby models or projects therapeutic effects on somatic instability' and / or cell survival across an affected patient population characterized by a distribution of repeat lengths and stages of intervention.

[0246] The described techniques (e.g., a computational model and related method(s)) for modeling or projecting therapeutic effect on SI and / or cell survival repeat expansion disorders and / or in genetic neurodegenerative disorders, w ith a focus on HD, predicts that for a hypothetical patient, presenting with a germline CAG length of 40, treated at age 30, a SI-targeting gene therapy with conservative efficacy characteristics achieves substantial therapeutic benefit: a 15 year delay in the onset of motor changes, and 12 and 15 year delays086013-0589445in the onset of HD-ISS Stage 2 and Stage 3, respectively. To evaluate the treatable range of HD patients, the model has been expanded to predict therapeutic benefit across a spectrum of germline CAG lengths and ages. This analysis predicts that a therapeutic could provide a minimum therapeutic benefit of > 5 years in delay to onset of motor symptoms for the majority7of HD patients and that it has the potential to confer the greatest benefit when delivered early, prior to the onset of motor symptoms. Another notable determination is that knockdown has a greater impact on therapeutic benefit than transduction consistent with the theoretical advantage of robust somatic instability reduction toward preserving even a limited number of MSNs.EXAMPLE 11Materials and MethodsSimulating CAG repeat expansion in MSNs

[0247] The simulation of CAG repeat expansion was built in R and works by first instantiating a population of MSNs (plotted as dots) each with a CAG repeat as an attribute. Each CAG repeat begins the simulation at year 0 at the germline CAG length. With each successive year in a simulated patient lifetime the CAG repeat attribute of each cell is modified in concordance with the Handsaker two-phase linear model of CAG repeat expansion p(x) = rl*max(x-Tl,0) + r2*max(x-T2,0) (2). The stochastic nature of CAG repeat expansion leads to variable expansion that compounds with longer repeats expanding more rapidly and cells with shorter repeats more likely to persist at shorter lengths. As CAG repeats expand their length may surpass 150 CAGs and enter a period of continuous transcriptional dysregulation. This 150 CAG length is used as an exemplary threshold for classifying cells as healthy (<=150) and dysregulated (>150). The rate of transduction and rate of SI lowering was parameterized and here is demonstrated the results of a simulated gene therapy application yielding a modest 50% MSH3 lowering in 50% of MSNs.Predicting therapeutic benefit

[0248] To predict therapeutic benefit, simulations are run as described herein in both the presence and absence of a SI modifying therapeutic. Then an HD landmark is selected for which a therapeutic benefit is calculated. Example landmarks include the onset of HD-ISS stage (I, II, or III), or CAP- 100. Then we identify the percentage of healthy cells remaining at the year the HD landmark of interest is reached in the untreated simulation. Then, by086013-0589445identifying the age at which the equivalent percentage of healthy cells is reached in the treated simulation, the difference in years between the untreated and treated simulations was calculated. This difference in age at which the HD landmark is reached is the predicted therapeutic benefit.AA V vector production

[0249] Recombinant AAV (rAAV) vectors were generated by the Research Vector Core at the Raymond G. Perelman Center for Cellular and Molecular Therapeutics at The Children’s Hospital of Philadelphia (CHOP). siSPOTR was used to design artificial miRNAs targeting human and NHP MSH3 (miMSH3) (23). miMSH3 were embedded in the pri-miRNA miR-30 or miR-451 scaffold and cloned into a rAAV plasmid shuttle with AAV2 inverted terminal repeat sequences and kanamycin selection (37). rAAV vectors were produced by the standard calcium phosphate transfection method in HEK293 cells with the AdHelper plasmid. AAV-DB-3 Rep2 / Capl peptide modified packaging plasmid, and rAAV shuttle plasmid with double CsCl purification (18, 38). Vector titers were determined by ddPCR.NHP procedures

[0250] All NHP procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals (National Research Council) and were approved by the CHOP Research Institute Animal Care and Use Committee. NHPs received MRI-guided bilateral GP injections with AAV test articles or vehicle control as indicated in Table 1. AAV formulation buffer (CHOP Research Vector Core) was used as a vehicle control and to dilute AAVs to the target concentration prior to dosing. Animals were sedated with ketamine and xylazine and maintained on isoflurane anesthesia during the procedure. The ClearPoint Preclinical Orchestra and navigation software were used for targeting. Injections were performed using a ClearPoint cannula (CUS-SMFL-03) attached to a syringe pump set to 1 uL / min. The two hemispheres were injected sequentially using the same cannula and an injection volume per hemisphere set to half of the total dose volume per animal in Table 1. A 10-minute dwell period was included following each injection before removing the cannula. Following both injections, the skin was sutured and the animal recovered, given buprenorphine SR as analgesia, and monitored postoperatively for pain and welfare.

[0251] After 4 weeks, NHPs were sedated and transcardially perfused with ice-cold saline. The brain was removed and sliced in 4-mm coronal slabs in a rhesus macaque brain matrix.086013-0589445For molecular analysis, brain regions of interest and other tissue types were collected and snap frozen in liquid nitrogen. For RNAscope FISH, brain slabs were postfixed in 10% neutral buffered formalin (NBF) for 24 hours, transferred to saline, and stored refrigerated for no more than 6 days before embedding in paraffin.RT-qPCRfor gene expression

[0252] Total RNA was extracted from NHP brain samples using the Quick-DNA / RNA Miniprep Plus kit (Zymo Research D7003) and quantified with the Qubit BR Assay kit (Life Technologies Q10210). cDNA was generated from 500 ng total RNA with Maxima H minus reverse transcriptase (Thermo Scientific EP0752) and random hexamers (Life Technologies SO 142). qPCR was performed with Luna Primer Probe 2X Master Mix (NEB M3004) including primer probe sets for rhesus MSH3 (Applied Biosystems Rh00989001_ml) and TBP (Applied Biosystems Rh00427620_ml). Total RNA was extracted from HdhQl 11 mouse brain samples with MagMAX mirV ana total RNA isolation kit (Thermo Scientific A27828) and KingFisher Flex system using a qualified RT-qPCR method (BioAgilytix Labs, Durham, NC). Tissue samples were combined with 300 pL of lysis binding mix (containing lysis buffer and 2-mercaptoethanol). The samples were homogenized using TissueLyser II (Qiagen) with 5 mm stainless steel beads at 25 Hz for 2 x 3-minute cycles. The RNA eluates were quantitated using Qiagen QIAxpert. RT-qPCR was performed using the TaqMan Fast Virus 1-Step Master Mix (Applied Biosystems 4444432) and duplexed with primers and probes for mouse Msh3 Mouse (Applied Biosystems Mm00487756_ml) and mouse ActB ACTB (Applied Biosystems Mm00607939_sl). Target gene Ct values were normalized to the housekeeping gene (ACt), then the average values were compared to average values of control samples (AACt). The 2('AAC9 method was used to calculate fold change in gene expression relative to vehicle controls.Stem-loop RT-qPCRfor miRNA expression

[0253] Total RNA was extracted from NHP brain samples using the Quick-DNA / RNA Miniprep Plus kit (Zymo D7003) according to the manufacturer’s protocol. RNA samples were quantified by Qubit BR Assay kit according to the manufacturer’s protocol (Invitrogen Q10210). Stem loop reverse transcription was carried out using TaqMan MicroRNA Reverse Transcription Kit, (Thermo Scientific 4366597 with 10 ng of RNA input and custom Taqrnan small RNA primer (Life Technologies 4398988). RT-qPCR was carried out using Luna086013-0589445Primer Probe 2X Master Mix (NEB M3004). To quantify absolute miRNA levels, an HPLC purified RNA oligonucleotide (Integrated DNA Technologies) corresponding to the miMSH3 guide strand was used to generate a standard curve from a range of le2 to le8 copies per reaction to interpolate miMSH3 expression levels. Data were normalized to miRNA copies per ug of total RNA input.MPCRfor A A V biodistribution

[0254] Genomic DNA (gDNA) was extracted from NHP brain samples using the Quick-DNA / RNA Miniprep Plus kit (Zymo Research D7003) according to the manufacturer's protocol and quantified using Qubit dsDNA BR assay kit (Invitrogen Q32853). gDNA input for the GP, caudate, and putamen were 0.66 or 6.6 ng per well. Wells with saturated positive droplets were excluded. When more than one concentration was tested and concentrations were in range, copy numbers were averaged. ddPCR were performed on the QX200 (BioRad) according to manufacturer’s instructions for probe-based assays with custom primers probe sets targeting the CAG promoter sequence (Integrated DNA Technologies) in the AAV genome and NHP RPP30 sequence (Integrated DNA Technologies) as a reference for normalization. AAV vector genome copies were normalized to NHP RPP30 copies to quantify AAV vector genomes per diploid genome using Bio-Rad QX Manager Software (v2.2.0.71).Capillary immunoassay

[0255] Tissues were lysed in RIPA buffer (Pierce 89900) with IX HALT protease inhibitor and homogenized using 5 mm stainless steel beads (Qiagen 69989) with Tissue Lyser LT at 50 Hz for 1 minute, incubated on ice for 30 seconds, and repeated for 3 cycles total, followed by incubation on ice for 5 minutes. Cell lysate debris was removed by centrifuging for 20 minutes at 16,000 g at 4 degrees Celsius, and supernatants were collected and stored in LoBind tubes at -80 degrees Celsius. Protein concentrations were quantified with Pierce BCA protein assay kit (ThermoFisher 23225). Samples were normalized to 2 mg / mL in 0. IX Sample Buffer and run on Jess capillary immunoassay (Protein Simple, SM-W001) following manufacturer's protocol. Primary mouse anti-MSH3 (BD Biosciences 611390) and secondary anti-mouse HRP (Protein Simple DM-002) were used for detection of MSH3 using default parameters Data was analyzed using Compass for SW software (v6.3, Protein Simple). Data are plotted as MSH3 peak areas normalized to vehicle controls.086013-0589445RNAscope FISH

[0256] NHP brain slabs were trimmed, embedding in paraffin, and microtome-sectioned at a thickness of 5 microns at StageBio (Frederick, MD). Unstained tissue sections mounted on slides were shipped to CHOP and stored refrigerated with desiccants prior to staining.Fluorescence in situ hybridization (FISH) was performed using the RNAscope multiplex fluorescent reagent kit v2 assay (Advanced Cell Diagnostics, Cat. #323100-USM) following the manufacturer’s instructions. RNAscope probes PPP1R1B (DARPP32,' Cat. 1241971-C3) and MSH3 (Cat. #1691821-C2) were used to detect mRNA transcripts in the putamen and caudate nucleus of rhesus macaques. Opal690 and Opal620 fluorophores (Akoya Biosciences, FP 1497001 KT and FP 1495001 KT) were used to detect the multiplexed PPP J RIB (DARPP32)-C3 and MSH3-C2 probes, respectively. Slides were imaged using a Leica SP8 confocal microscope with LAS X v.3.7 software.Automated cell counting

[0257] Quantification of cells from fluorescence RNAscope FISH images was performed using Qupath. Cells were segmented using the “WatershedCellDetection"’ function. Nuclei were defined using the nuclear stain channel and, due to lack of a cytoplasmic stain, cytoplasmic regions were added using a lOuM “cellExpansion” region around each nucleus. Subcellular MSH3 puncta were quantified using the “SubcellularDetection” function. A custom groovy script w as used to designate each subcellular detection as nuclear or cytoplasmic. A counts table was exported containing these per cell puncta counts and additional measurements. Data management and visualization was performed using a custom R script.HdhQlll mice

[0258] HdhQl 11 KI heterozygous mice w ere previously described and bred at The Jackson Laboratory (Bar Harbor, ME) (39). Procedures were approved by the Institutional Animal Care and Use Committee in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals. Animals were balanced into treatment groups by body weight. At 8 weeks of age, test articles were administered to animals in groups 4-7 or vehicle in group 3 via simultaneous bilateral striatal injections. Standard aseptic surgical procedures were used to perform the injections. All injections were performed under isoflurane anesthesia (3-4% induction, 1-2% maintenance). The stereotaxic coordinates were 0.86 mm086013-0589445anterior and ±1.8 mm lateral with respect to bregma, and -2.5 mm dorsal ventral from the dura, and 5 pL volume per hemisphere was injected into the striatum at a 0.2 pL / minute infusion rate. Following infusion, the needles were left in place for 5 minutes to allow the test article to diffuse, the needle was then slowly retracted over 1-2 minutes. The skin was sutured over the injection site and the animal was returned to the home cage.Somatic instability index

[0259] Somatic instability testing was performed at Transnetyx (Culver City, CA).Genomic DNA was extracted using a proprietary magnetic bead-based extraction protocol. CAG repeat PCR amplification of the Htt gene containing CAG repeats was amplified using a Thermo Hot Start Master Mix and a set of primers, one of which was labeled with 6-FAM fluorescent dye. PCR amplification was conducted on an Simpliamp thermocycler using a touchdown PCR protocol to improve specificity and yield. Fragment analysis and sizing were performed by PCR products (2 pL) mixed with 0.5 pL of ROX1500 size standard and 7.5 pL of Hi-Di formamide. Samples were first normalized to ensure consistent input of 20 ng of DNA into the reaction. Samples are denatured at 96°C for 2 minutes and analyzed on an ABI 3730XL capillary electrophoresis platform. Data was processed using GeneMapper 5.0 software. CAG somatic instability index calculation was done using a Python script based on the method from Jong-Min Lee et al. (2010). Peak heights >1000 RFU are required for reliable instability index calculation. 10% thresholds were used for striatum and 5% thresholds were used for cortex.Statistical analysis

[0260] Differences between untreated and treated groups were compared using one-way ANOVA with Dunnetf s test for multiple comparisons for mouse Msh3 mRNA levels and somatic instability index. Differences between groups were considered to be significant at a / ’ value of < 0.05. All results are shown as the mean ± SD. Statistical analyses were performed with GraphPad Prism vlO.Dual-Luciferase screening ofmiMSH3 sequences

[0261] Twenty-nine miMSH3 guide strand sequences w ere embedded in the miR-30 scaffold and cloned into the 3 ' UTR of EGFP in the pCl mammalian expression plasmid (Promega El 731). Corresponding miMSH3 target sequences were cloned into the 3’ UTR of086013-0589445Renilla luciferase of the psiCHECK-2 plasmid (Promega C8021), which also expresses Firefly luciferase for normalization. HEK293 cells were co-transfected with both plasmids using Lipofectamine 2000. After 24 hours, cells were lysed and luminescence was measured with the Dual-Luciferase reporter assay system (Promega El 910) using the GloMax Discover plate reader (Promega). Renilla luciferase was normalized to firefly luciferase to quantify miMSH3 guide strand loading and knockdown efficiency relative to miSAFE, a non-targeting control miRNA (41). miHDSl, a miRNA targeting HIT. was included as positive control (41, 42). Passenger strand loading was assessed using the psiCHECK-2 plasmid containing antisense target sequence in the 3’ UTR of Renilla luciferase.RNA sequencing

[0262] RNA sequencing was performed at GENEWIZ (South Plainfield, NJ). RNA quantify and integrity were assessed using Qubit 4.0 Fluorometer (Thermo Scientific) and 4200 TapeStation (Agilent Technologies). Small RNA sequencing libraries were generated with NEB Small RNA library Prep Kit (New England Biolabs). Illumina 3’ and 5’ adapter were ligated to 100 ng of RNA molecules with a 5 ’-phosphate and a 3 ’-hydroxyl group sequentially, followed by reverse transcription and PCR amplification with indexed primers. Amplified cDNA (-145 - 160 bp) was purified by PAGE, gel-extracted, and concentrated by ethanol precipitation. mRNA sequencing libraries were prepared using the NEBNext Ultra II RNA Library Prep Kit (New England Biolabs) for Illumina following the manufacturer's protocol. Poly (A) enrichment from 100 ng total RNA was performed with oligo d(T) beads, mRNA was fragmented (15 min, 94 °C), and converted to double-strand. End-repair. adenylated at 3'ends. adapter ligation, and PCR indexing were performed with limited cycles. Final libraries were confirmed by TapeStation (Agilent Technologies) and quantified by Qubit 4.0 Fluorometer (Thermo Scientific) as well as by quantitative PCR (qPCR) (KAPA Biosystems). Libraries were clustered and sequenced (2x150 bp paired-end; >80% bases >Q30) on the Illumina NovaSeq instrument according to manufacturer’s instructions. Demultiplexing was performed using Illumina's bcl2fastq 2.20 software with a tolerance of one mismatch per index. miRNA off target analysis was performed using DESeq2 differential expression analysis. miRNA processing fidelity was assessed using a custom R script.RT-qPCR for gene expression (cells and wildtype mice)086013-0589445

[0263] Total RNA was extracted from HEK293 cells, N2a cells, and mouse brain tissue using TRIzol (Invitrogen 15596026) according to the manufacturer's protocol. Glycoblue (Invitrogen AM9515) was added to the aqueous phase in the isopropanol precipitation step, and a single wash was performed with cold 75% ethanol. RNA was DNase treated (TURBO DNA-free kit, Invitrogen AMI 9107) followed by reverse transcription using 500 ng of RNA with the High-Capacity cDNA Reverse Transcription Kit and random primers (Applied Biosystems 4368814). qPCR probes used for HEK293 cells were human GAPDEI (Applied Biosystems Hs99999905_ml) and human MSH3 (Integrated DNA Technologies Hs.PT.58.46270796). The probes used for N2a cell and mouse tissue samples were mouse ACTB (Applied Biosystems Mm00607939_sl) and mouse MSH3 (Applied Biosystems Mm0487756 ml). qPCR was performed with Taqman Universal Master Mix (Applied Biosystems 4304437) with cycling conditions, 95°C for 10 min., 40 cycles of 95°C for 10 s, and 60°C for 1 min. Relative expression levels were determined using the 2(-AACt)method with GAPDH or ACTB as reference genes for normalization. miR-30.miHDSl and miR-451.miH12, miRNAs targeting HIT. were included as positive controls (41-43).qPCR for AA V biodistribution (NHP and mouse)

[0264] AAV genome biodistribution was quantified in NHP and mouse brain and peripheral tissues using qualified qPCR methods (Northern Bio, Portage MI). Tissues were weighed and homogenized in lysis buffer with ceramic beads at 6000 rpm for 30 s using a Precellys tissue homogenizer. Lysates were processed with the QIASymphony (QIAGEN) for automated genomic DNA (gDNA) extraction using the QIASymphony DSP DNA mini kit (QIAGEN). DNA concentration and purity were assessed with NanoDrop 8000 spectrophotometer. qPCR was performed using TaqMan Fast Advanced master mix (Applied Biosystems) with a custom primer-probe set (Applied Biosystems) targeting the CAG promoter and 5’ miR-30 region in the AAV genome. Reactions were run on the QuantStudio 7 Flex PCR system using standard cycling conditions. A linearized plasmid DNA standard curve (25 to 108copies) was used to interpolate vector copy numbers. Copy number per microgram of gDNA was converted to copies per diploid genome using the host species genome size (bp).

[0265] Continuing with the remaining figures, FIG.7 illustrates a system 710 comprising a modelling engine 712 and other components configured to facilitate modeling or projecting therapeutic effect on somatic instability7and / or cell survival in HD. Modelling engine 712 is086013-0589445executed by one or more of the computers described below with reference to FIG. 8 and may include an application program interface (API) server 726, a web server 728, a data store 730, and a cache server 732. These components, in some embodiments, communicate with one another in order to provide the functionality of modelling engine 712. Data store 730 may store any data that facilitates the method(s) described herein. The cache server 732 may expedite access to this data by storing likely relevant data in relatively high-speed memory, for example, in random-access memory or a solid-state drive. The web server 728 may serve webpages having graphical user interfaces that facilitate transfer of information to and / or from users, or other displays. The API server 726 may serve data to various applications that process data related to the modelling described herein. The operation of these components 726, 728. and 730 may be coordinated by a controller 714, which may bidirectionally communicate with each of these components or direct the components to communicate with one another. Communication may occur by transmitting data between separate computing devices (e.g., via transmission control protocol / intemet protocol (TCP / IP) communication over a network), by transmitting data between separate applications or processes on one computing device; or by passing values to and from functions, modules, or objects within an application or process, e.g., by reference or by value.

[0266] Interaction with users may occur via a website or a native application viewed on a desktop computer, tablet, or a laptop of the user. And in some cases, such interaction occurs via a mobile website viewed on a smart phone, tablet, or other mobile user device, or via a special-purpose native application executing on a smart phone, tablet, or other mobile user device. Facilitating interactions across a variety of devices is expected to make it easier for a user to execute the model where convenient for the user. To illustrate an example of the environment in which the modelling engine 712 operates, the illustrated embodiment of FIG.7 includes a number of components with which modelling engine 712 communicates: mobile user devices 734 and 736; a desk-top user device 738; and external resources 746. Each of these devices communicates with modelling engine 712 via a network 750. such as the Internet or the Internet in combination with various other networks, like local area networks, cellular networks, or personal area networks.

[0267] The mobile user devices 734 and 736 may be smart phones, tablets, gaming devices, or other hand-held networked computing devices having a display, a user input device (e.g., buttons, keys, voice recognition, or a single or multi-touch touchscreen), memory (such as a086013-0589445tangible, machine-readable, non-transitory memory), a network interface, a portable energy source (e.g., a battery ), and a processor (a term which, as used herein, includes one or more processors) coupled to each of these components. The memory of the mobile user devices 734 and 736 may store instructions that when executed by the associated processor provide an operating system and various applications, including a web browser 742 or a native mobile application 740. The desk-top user device 738 may also include a web browser 744. In addition, the desktop user device 738 may include a monitor; a keyboard; a mouse; memory; a processor; and a tangible, non-transitory, machine-readable memory storing instructions that when executed by the processor provide an operating system and the web browser. The native application 740 and the web browsers 742 and 744, in some embodiments, are operative to provide a graphical user interface that communicates with modelling engine 712 and facilitates user interaction with data from modelling engine 712. The web browsers 742 and 744 may be configured to receive a website from modelling engine 712 having data related to instructions (for example, instructions expressed in JavaScriptTM) that when executed by the browser (which is executed by the processor) cause the mobile user device 736 and / or the desktop user device 738 to communicate with modelling engine 712 and facilitate user interaction with data from modelling engine 712. The native application 740 and the web browsers 742 and 744, upon rendering a webpage and / or a graphical user interface from modelling engine 712, may generally be referred to as client applications of modelling engine 712, which in some embodiments may be referred to as a server.Embodiments, however, are not limited to client / server architectures, and modelling engine 712, as illustrated, may include a variety of components other than those functioning primarily as a server. Three user devices are shown, but embodiments are expected to interface with substantially more, with more than 100 concurrent sessions and serving more than 1 million users distributed over a relatively large geographic area, such as a state or the entire United States.

[0268] External resources 746, in some embodiments, include sources of information such as databases, websites, etc.; external entities participating with system 710 (e.g., systems or networks associated with sources of data), one or more servers outside of system 710, a network (e.g., the internet), electronic storage, equipment related to Wi-Fi ™ technology7, equipment related to Bluetooth® technology, data entry devices, or other resources. In some implementations, some or all of the functionality attributed herein to external resources 746 may be provided by resources included in system 710. External resources 746 may be086013-0589445configured to communicate with modelling engine 712, mobile user devices 734 and 736, desktop user device 738. and / or other components of system 710 via wired and / or wireless connections, via a network (e.g., a local area network and / or the internet), via cellular technology, via Wi-Fi technology, and / or via other resources.

[0269] Thus, modelling engine 712, in some embodiments, operates in the illustrated environment by communicating with a number of different devices and transmitting instructions to various devices to communicate with one another. The number of illustrated external resources 746, desktop user devices 738, and mobile user devices 736 and 734 is selected for explanatory purposes only, and embodiments are not limited to the specific number of any such devices illustrated by FIG. 7. which is not to imply that other descriptions are limiting.

[0270] Modelling engine 712 of some embodiments includes a number of components introduced above that facilitate interaction by users. For example, the illustrated API server 726 may be configured to communicate data about cells or cell collections, and / or other information via a protocol, such as a representational-state-transfer (REST)-based API protocol over hypertext transfer protocol (HTTP) or other protocols. Examples of operations that may be facilitated by the API server 726 include requests to display, link, modify, add, or retrieve portions or all of such data, or other information. API requests may identify which data is to be displayed, linked, modified, added, or retrieved by specifying criteria for identify ing records, such as queries for retrieving or processing information about a particular cell type, for example. In some embodiments, the API server 726 communicates with the native application 740 of the mobile user device 734 or other components of system 710.

[0271] The illustrated web server 728 may be configured to display, link, modify, add, or retrieve portions or all of a dataset, or other information encoded in a webpage (e.g. a collection of resources to be rendered by the browser and associated plug-ins, including execution of scripts, such as JavaScriptTM, invoked by the webpage). In some embodiments, the graphical user interface presented by the webpage may include inputs by which the user may enter or select data, such as clickable or touchable display regions or display regions for text input. Such inputs may prompt the browser to request additional data from the web server 728 or transmit data to the web server 728, and the web server 728 may respond to such requests by obtaining the requested data and returning it to the user device or acting upon the transmitted data (e.g., storing posted data or executing posted commands). In some086013-0589445embodiments, the requests are for a new webpage or for data upon which client-side scripts will base changes in the webpage, such as XMLHttpRequest requests for data in a serialized format, e.g. JavaScriptTM object notation (JSON) or extensible markup language (XML). The web server 728 may communicate with web browsers, such as the web browser 742 or 744 executed by user devices 736 or 738. In some embodiments, the webpage is modified by the web server 728 based on the type of user device, e.g., with a mobile webpage having fewer and smaller images and a narrower width being presented to the mobile user device 736, and a larger, more content rich webpage being presented to the desk-top user device 738. An identifier of the type of user device, either mobile or non-mobile, for example, may be encoded in the request for the webpage by the web browser (e.g., as a user agent type in an HTTP header associated with a GET request), and the web server 728 may select the appropriate interface based on this embedded identifier, thereby providing an interface appropriately configured for the specific user device in use.

[0272] The illustrated data store 730 stores some or all of the data on which the modelling described herein is based. The data store 730 may include various types of data stores, including relational or non-relational databases, document collections, hierarchical key -value pairs, or memory images, for example. Such components may be formed in a single database, document, or the like, or may be stored in separate data structures. In some embodiments, the data store 730 comprises electronic storage media that electronically stores information. The electronic storage media of data store 730 may include one or both of system storage that is provided integrally (i.e., substantially non-removable) with system 710 and / or removable storage that is removably connectable to the system 710 via, for example, a port (e.g., a USB port, a firewire port, etc.) or a drive (e.g., a disk drive, etc.). The data store 730 may be (in whole or in part) a separate component within system 710, or the data store 730 may be provided (in whole or in part) integrally with one or more other components of system 710 (e.g., processors 716, etc.). In some embodiments, the data store 730 may be located in a data center, in a server that is part of external resources 746, in a computing device 734, 736, or 738, or in other locations. The data store 730 may include one or more of optically readable storage media (e.g., optical disks, etc.), magnetically readable storage media (e.g., magnetic tape, magnetic hard drive, floppy drive, etc.), electrical charge-based storage media (e.g., EPROM, RAM, etc.), solid-state storage media (e.g., flash drive, etc.), or other electronically readable storage media. The data store 730 may store software algorithms, information determined by the controller 714, information received via the graphical user interface086013-0589445displayed on computing devices 734, 736, and / or 738, information received from external resources 746, or other information accessed by system 710 to function as described herein.

[0273] The controller 714 is configured to coordinate the operation of the other components of modelling engine 712 to provide the functionality described herein.Controlled components may include one or more processors 716, which are configured to execute instructions 718, and / or other components. The controller 714 may be configured to direct the operation of such components by software; hardware; firmware; some combination of software, hardware, or firmware; or other mechanisms for configuring processing capabilities.

[0274] FIG. 8 is a diagram that illustrates an exemplary computing system 800.Computing system 800 may execute some or all of the model described herein, for example. Various portions of the methods described herein, may include or be executed on one or more computer systems the same as or similar to computing system 800. For example, computing engine 12, mobile user device 34, mobile user device 36, desktop user device 38, external resources 46 and / or other components of the system 10 (FIG. 7) may be and / or include one more computer systems the same as or similar to computing system 800. Further, processes, modules, processor components, and / or other components of the system 10 described herein may be executed by one or more processing systems similar to and / or the same as that of computing system 800.

[0275] Computing system 800 may include one or more processors (e.g.. processors 810a-810n) coupled to system memory 820, an input / output I / O device interface 830, and a network interface 840 via an input / output (I / O) interface 850. A processor may include a single processor or a plurality of processors (e.g., distributed processors). A processor may be any suitable processor capable of executing or otherwise performing instructions. A processor may include a central processing unit (CPU) that carries out program instructions to perform the arithmetical, logical, and input / output operations of computing system 800. A processor may execute code (e.g., processor firmware, a protocol stack, a database management system, an operating system, or a combination thereof) that creates an execution environment for program instructions. A processor may include a programmable processor. A processor may include general or special purpose microprocessors. A processor may receive instructions and data from a memory (e.g., system memory 820). Computing system 800 may be a uniprocessor system including one processor (e.g., processor 810a), or a multi-processor system086013-0589445including any number of suitable processors (e.g., 810a-810n). Multiple processors may be employed to provide for parallel or sequential execution of one or more portions of the techniques described herein. Processes, such as logic flows, described herein may be performed by one or more programmable processors executing one or more computer programs to perform functions by operating on input data and generating corresponding output. Processes described herein may be performed by, and apparatus can also be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit). Computing system 800 may include a plurality of computing devices (e.g., distributed computer systems) to implement various processing functions.

[0276] I / O device interface 830 may provide an interface for connection of one or more I / O devices 860 to computer system 800. I / O devices may include devices that receive input (e.g., from a user) or output information (e.g., to a user). I / O devices 860 may include, for example, graphical user interface presented on displays (e.g., a cathode ray tube (CRT) or liquid crystal display (LCD) monitor, touch screens, etc.), pointing devices (e.g., a computer mouse or trackball), keyboards, keypads, touchpads, scanning devices, voice recognition devices, gesture recognition devices, printers, audio speakers, microphones, cameras, or the like. I / O devices 860 may be connected to computer system 800 through a wired or wireless connection. I / O devices 860 may be connected to computer system 800 from a remote location. I / O devices 860 located on remote computer system, for example, may be connected to computer system 800 via a network and network interface 840.

[0277] Network interface 840 may include a network adapter that provides for connection of computer system 800 to a network. Network interface may 840 may facilitate data exchange between computer system 800 and other devices connected to the network.Network interface 840 may support wired or wireless communication. The network may include an electronic communication network, such as the Internet, a local area network (LAN), a wide area netw ork (WAN), a cellular communications netw ork, or the like.

[0278] System memory' 820 may be configured to store program instructions 870 or data 880. Program instructions 870 may be executable by a processor (e.g., one or more of processors 810a-810n) to implement one or more embodiments of the present techniques. Instructions 870 may include modules and / or components of computer program instructions for implementing one or more techniques described herein with regard to various processing086013-0589445modules and / or components. Program instructions may include a computer program (which in certain forms is known as a program, software, software application, script, or code). A computer program may be written in a programming language, including compiled or interpreted languages, or declarative or procedural languages. A computer program may include a unit suitable for use in a computing environment, including as a stand-alone program, a module, a component, or a subroutine. A computer program may or may not correspond to a file in a file system. A program may be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program may be deployed to be executed on one or more computer processors located locally at one site or distributed across multiple remote sites and interconnected by a communication network.

[0279] System memory 820 may include a tangible program carrier having program instructions stored thereon. A tangible program carrier may include a non-transitory computer readable storage medium. A non-transitory computer readable storage medium may include a machine readable storage device, a machine readable storage substrate, a memory device, or any combination thereof. Non-transitory computer readable storage medium may include non-volatile memory (e.g.. flash memory, ROM, PROM, EPROM, EEPROM memory), volatile memory (e.g., random access memory (RAM), static random access memory (SRAM), synchronous dynamic RAM (SDRAM)), bulk storage memory (e.g., CD-ROM and / or DVD-ROM, hard-drives), or the like. System memory7820 may include a non-transitory computer readable storage medium that may have program instructions stored thereon that are executable by a computer processor (e.g.. one or more of processors 810a-810n) to cause the subject matter and the functional operations described herein. A memory (e.g., system memory 820) may include a single memory device and / or a plurality of memory devices (e.g., distributed memory7devices). Instructions or other program code to provide the functionality described herein may be stored on a tangible, non-transitory computer readable media. In some cases, the entire set of instructions may be stored concurrently on the media, or in some cases, different parts of the instructions may be stored on the same media at different times, e.g., a copy may be created by writing program code to a first-in-first-out buffer in a network interface, where some of the instructions are pushed out of the buffer before other portions of the instructions are written to the buffer, with all of the instructions residing in memory on the buffer, just not all at the same time.086013-0589445

[0280] I / O interface 850 may be configured to coordinate I / O traffic between processors 810a-810n. system memory 820, network interface 840, I / O devices 860. and / or other peripheral devices. I / O interface 850 may perform protocol, timing, or other data transformations to convert data signals from one component (e.g., system memory 820) into a format suitable for use by another component (e.g., processors 810a-810n). I / O interface 850 may include support for devices attached through various types of peripheral buses, such as a variant of the Peripheral Component Interconnect (PCI) bus standard or the Universal Serial Bus (USB) standard.

[0281] Embodiments of the techniques described herein may be implemented using a single instance of computer system 800 or multiple computer systems 800 configured to host different portions or instances of embodiments. Multiple computer systems 800 may provide for parallel or sequential processing / execution of one or more portions of the techniques described herein.

[0282] Those skilled in the art will appreciate that computer system 800 is merely illustrative and is not intended to limit the scope of the techniques described herein.Computer system 800 may include any combination of devices or software that may perform or otherwise provide for the performance of the techniques described herein. For example, computer system 800 may include or be a combination of a cloud-computing system, a data center, a server rack, a server, a virtual server, a desktop computer, a laptop computer, a tablet computer, a server device, a client device, a mobile telephone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a vehicle-mounted computer, a television or device connected to a television (e.g., Apple TV ™), or a Global Positioning System (GPS), or the like. Computer system 800 may also be connected to other devices that are not illustrated, or may operate as a stand-alone system. In addition, the functionality provided by the illustrated components may in some embodiments be combined in fewer components or distributed in additional components. Similarly, in some embodiments, the functionality of some of the illustrated components may not be provided or other additional functionality may be available.

[0283] Those skilled in the art will also appreciate that while various items are illustrated as being stored in memory or on storage while being used, these items or portions of them may be transferred between memory and other storage devices for purposes of memory management and data integrity. Alternatively, in other embodiments some or all of the086013-0589445software components may execute in memory on another device and communicate with the illustrated computer system via inter-computer communication. Some or all of the system components or data structures may also be stored (e.g., as instructions or structured data) on a computer-accessible medium or a portable article to be read by an appropriate drive, various examples of which are described above. In some embodiments, instructions stored on a computer-accessible medium separate from computer system 800 may be transmitted to computer system 800 via transmission media or signals such as electrical, electromagnetic, or digital signals, conveyed via a communication medium such as a network or a wireless link. Various embodiments may further include receiving, sending, or storing instructions or data implemented in accordance with the foregoing description upon a computer-accessible medium. Accordingly, the present invention may be practiced with other computer system configurations.

[0284] The present systems and methods may be used in human medical applications. Suitable subjects therefore include mammals, such as humans, as well as non-human mammals. Human subjects include fetal, neonatal, infant juvenile and adult (e.g., HD) subjects.

[0285] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention belongs. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the invention, suitable methods and materials are described herein.

[0286] All patents, patent applications, publications, and other references cited herein are incorporated by reference in their entirety. In case of conflict, the specification, including definitions, will control.

[0287] All of the features disclosed herein may be combined in any combination. Each feature disclosed in the specification may be replaced by an alternative feature serving a same, equivalent, or similar purpose. Thus, unless expressly stated otherwise, disclosed features are an example of a genus of equivalent or similar features.

[0288] The techniques described above are disclosed using affirmative language to describe the numerous embodiments. These techniques also specifically include embodiments in which particular subject matter is excluded, in full or in part, such as substances or materials,086013-0589445method steps and conditions, protocols, or procedures. For example, in certain embodiments, method steps are excluded.

[0289] A number of embodiments have been described. Nevertheless, one skilled in the art, without departing from the spirit and scope of the description, can make various changes and modifications to adapt these techniques to various usages and conditions. Accordingly, the following examples are intended to illustrate but not limit the scope of the claimed techniques in any way.REFERENCES AND NOTES1. L. Kennedy, E. Evans, C. M. Chen, L. Craven, P. J. Detloff, M. Ennis, P. F.Shelboume, Dramatic tissue-specific mutation length increases are an early molecular event in Huntington disease pathogenesis. Hum Mol Genet 12, 3359-3367 (2003).2. R. E. Handsaker, S. Kashin, N. M. Reed, S. Tan. W. S. Lee, T. M. McDonald. K. Morris, N. Kamitaki, C. D. Mullally, N. R. Morakabati, M. Goldman, G. Lind, R. Kohli, E. Lawton, M. Hogan, K. Ichihara, S. Berretta. S. A. McCarroll, Long somatic DNA-repeat expansion drives neurodegeneration in Huntington's disease. Cell 188, 623-639 e619 (2025).3. J. H. Cha, Transcriptional dysregulation in Huntington's disease. Trends Neurosci 23, 387-392 (2000).4. R. I. Scahill, M. Farag, M. J. Murphy, N. Z. Hobbs, M. Leocadi, C. Langley, H. Knights, M. Ciosi, K. Fayer, M. Nakajima, O. Thackeray, J. Gobom, J. Ronnholm, S. Weiner, Y. R. Hassan, N. K. P. Ponraj, C. Estevez-Fraga, C. S. Parker, I. B. Malone, H. Hyare, J. D. Long, A. Heslegrave, C. Sampaio, H. Zhang, T. W. Robbins, H. Zetterberg, E. J. Wild, G. Rees, J. B. Rowe, B. J. Sahakian, D. G. Monckton, D. R. Langbehn, S. J. Tabrizi, Somatic CAG repeat expansion in blood associates with biomarkers of neurodegeneration in Huntington's disease decades before clinical motor diagnosis. Nat Med 31, 807-818 (2025).5. K. Matlik, M. Baffuto, L. Kus, A. L. Deshmukh, D. A. Davis, M. R. Paul. T. S.Carroll, M. C. Caron, J. Y. Masson, C. E. Pearson, N. Heintz, Cell-type-specific CAG repeat expansions and toxicity of mutant Huntingtin in human striatum and cerebellum. Nat Genet 56, 383-394 (2024).6. C. Pressl, K. Matlik, L. Kus, P. Darnell, J. D. Luo, M. R. Paul, A. R. Weiss, W.Liguore, T. S. Carroll, D. A. Davis, J. McBride, N. Heintz, Selective vulnerability of layer 5a corticostriatal neurons in Huntington's disease. Neuron 112, 924-941 e910 (2024).7. G. M. o. H. s. D. GeM-HD, Consortium. CAG Repeat Not Polyglutamine Length Determines Timing of Huntington's Disease Onset. Cell 178, 887-900 e814 (2019).8. D. J. H. Moss, A. F. Pardinas, D. Langbehn, K. Lo, B. R. Leavitt, R. Roos, A. Durr, S. Mead, T.-H. investigators, R. investigators, P. Holmans, L. Jones, S. J. Tabrizi, Identification of genetic variants associated with Huntington's disease progression: a genome-wide association study. Lancet Neurol 16, 701-711 (2017).9. G. M. o. H. s. D. C. GeM-HD, Identification of Genetic Factors that Modify Clinical Onset of Huntington's Disease. Cell 162, 516-526 (2 15).086013-058944510. B. McAllister, J. Donaldson, C. S. Binda, S. Powell, U. Chughtai, G. Edwards, J. Stone, S. Lobanov, L. Elliston, L. N. Schuhmacher, E. Rees, G. Menzies, M. Ciosi, A.Maxwell, M. J. Chao, E. P. Hong, D. Lucente, V. Wheeler, J. M. Lee, M. E. MacDonald, J. D. Long, E. H. Aylward, G. B. Landwehrmeyer, A. E. Rosser, R. I. o. t. E. H. s. d. network, J. S. Paulsen, P.-H. I. o. t. H. S. Group, N. M. Williams, J. F. Gusella, D. G. Monckton, N. D. Allen, P. Holmans, L. Jones, T. H. Massey, Exome sequencing of individuals with Huntington's disease implicates FAN1 nuclease activity in slowing CAG expansion and disease onset. Nat Neurosci 25, 446-457 (2022).11. E. L. Bunting, J. Donaldson, S. A. Cumming, J. Olive, E. Broom, M. Miclaus, J. Hamilton, M. Tegtmeyer, H. T. Zhao, J. Brenton, W. S. Lee. R. E. Handsaker, S. Li, B. Ford, M. Ryten, S. A. McCarroll, H. B. Kordasiewicz, D. G. Monckton, G. Balmus, M. Flower, S. J. Tabrizi, Antisense oligonucleotide-mediated MSH3 suppression reduces somatic CAG repeat expansion in Huntington's disease iPSC-derived striatal neurons. Sci TranslMed 17 , eadn4600 (2025).12. E. Dragileva, A. Hendricks, A. Teed, T. Gillis, E. T. Lopez, E. C. Friedberg, R.Kucherlapati, W. Edelmann, K. L. Lunetta, M. E. MacDonald, V. C. Wheeler, Intergenerational and striatal CAG repeat instability in Huntington's disease knock-in mice involve different DNA repair genes. Neurobiol Dis 33, 37-47 (2009).13. R. Ferguson, S. J. Tabrizi, Can MSH3 lowering stop HTT repeat expansion in its CAG tract? Afo / TTzer 31, 1509-1511 (2023).14. D. O'Reilly, J. Belgrad. C. Ferguson. A. Summers. E. Sapp, C. McHugh, E. Mathews, A. Boudi, J. Buchwald, S. Ly, D. Moreno, R. Furgal, E. Luu, Z. Kennedy, V. Hariharan, K. Monopoli, X. W. Yang, J. Carroll, M. DiFiglia, N. Aronin, A. Khvorova, Di-valent siRNA-mediated silencing of MSH3 blocks somatic repeat expansion in mouse models of Huntington's disease. Mol Ther 31, 3355-3356 (2023).15. R. Driscoll, L. Hampton, N. A. Abraham, J. D. Larigan, N. F. Joseph, J. C.Hernandez-Vega, S. Geisler, F. C. Yang, M. Deninger, D. T. Tran, N. Khatri, B. Godinho, G. A. Kinberger, D. R. Montagna, W. D. Hirst, C. L. Guardado, K. E. Glajch, H. M. Arnold, C. L. Gallant-Behm, A. Weihofen, Dose-dependent reduction of somatic expansions but not Htt aggregates by di-valent siRNA-mediated silencing of MSH3 in HdhQlll mice. Sci Rep 14, 2061 (2024).’16. R. Mouro Pinto, R. Murtha, A. Azevedo, C. Douglas, M. Kovalenko, J. Ulloa, S. Crescenti, Z. Burch, E. Oliver, M. Kesavan, S. Shibata, A. Vitalo, E. Mota-Silva, M. J. Riggs, K. Correia. E. Elezi, B. Demelo, J. B. Carroll, T. Gillis, J. F. Gusella, M. E. MacDonald, V. C. Wheeler, In vivo CRISPR-Cas9 genome editing in mice identifies genetic modifiers of somatic CAG repeat instability in Huntington's disease. Nat Genet 57, 314-322 (2025). 17. N. Wang, S. Zhang, P. Langfelder, L. Ramanathan, F. Gao, M. Plascencia, R. Vaca, X. Gu, L. Deng, L. E. Dionisio, H. Vu, E. Maciejewski, J. Ernst, B. C. Prasad, T. F. Vogt, S. Horvath, J. S. Aaronson, J. Rosinski, X. W. Yang, Distinct mismatch-repair complex genes set neuronal C AG-repeat expansion rate to drive selective pathogenesis in HD mice. Cell 188, 1524-1544 el522 (2025).18. D. E. Leib, Y. H. Chen, L. Tecedor, P. T. Ranum, M. S. Keiser, B. C. Lewandowski, E. M. Carrell. S. Arora, I. Huerta-Ocampo, D. Lai, C. M. Fluta, C. Cheng. X. Liu. B. L.Davidson, Optimized AAV capsids for basal ganglia diseases show robust potency and distribution. Nat Commun 16, 4653 (2025).086013-058944519. S. J. Tabrizi, S. Schobel, E. C. Gantman, A. Mansbach, B. Borowsky, P.Konstantinova, T. A. Mestre, J. Panagoulias, C. A. Ross, M. Zauderer, A. P. Mullin, K.Romero, S. Sivakumaran, E. C. Turner, J. D. Long, C. Sampaio, C. Huntington's Disease Regulatory Science, A biological classification of Huntington's disease: the Integrated Staging System. Lancet Neurol 21, 632-644 (2022).20. S. A. Schobel, G. Palermo. P Auinger. J. D. Long. S. Ma, O. S. Khwaja, D. Trundell. M. Cudkowicz, S. Hersch, C. Sampaio, E. R. Dorsey, B. R. Leavitt, K. D. Kieburtz, J. J. Sevigny, D. R. Langbehn, S. J. Tabrizi, C. C.-H. Track-Hd, C. H. S. G. Investigators. Motor, cognitive, and functional declines contribute to a single progressive factor in early HD.Neurology 89, 2495-2502 (2017).21. S. Lee, B. Weisburd, J. Lee, K. Correia, S. Zeng, S. S. Park, J. W. Shin, D. E. Choi, K. H. Kim, J. H. Jang, T. Gillis, H. L. Rehm, J. F. Gusella, M. E. MacDonald, J. M. Lee, Significant underascertainment in Huntington's disease. Brain Commun 7, fcafl94 (2025). 22. R. M. Spengler, X. Zhang, C. Cheng, J. M. McLendon, J. M. Skeie, F. L. Johnson, B. L. Davidson, R. L. Boudreau, Elucidation of transcriptome-wide microRNA binding sites in human cardiac tissues by Ago2 HITS-CLIP. Nucleic Acids Res 44, 7120-7131 (2016).23. R. L. Boudreau, R. M. Spengler, R. H. Hylock, B. J. Kusenda, H. A. Davis, D. A. Eichmann, B. L. Davidson, siSPOTR: a tool for designing highly specific and potent siRNAs for human and mouse. Nucleic Acids Res 41, e9 (2013).24. J. He, M. Kleyman, J. Chen, A. Alikaya, K. M. Rothenhoefer, B. E. Ozturk, M.Wirthlin, A. C. Bostan, K. Fish, L. C. Byrne, A. R. Pfenning, W. R. Stauffer, Transcriptional and anatomical diversity of medium spiny neurons in the primate striatum. Curr Biol 31, 5473-5486 e5476 (2021).25. T. C. Roberts, The MicroRNA Biology of the Mammalian Nucleus. Mol Ther Nucleic Acids 3, el 88 (2014).26. L. Kennedy, P. F. Shelboume, Dramatic mutation instability in HD mouse striatum: does poly glutamine load contribute to cell-specific vulnerability in Huntington's disease? Hum Mol Genet 9, 2539-2544 (2000).27. S. G. Aldous, E. J. Smith, C. Landles, G. F. Osborne, M. Canibano-Pico, I. M. Nita, J. Phillips, Y. Zhang, B. Jin, M. B. Hirst, C. L. Benn, B. C. Bond, W. Edelmann, J. R. Greene, G. P. Bates, A CAG repeat threshold for therapeutics targeting somatic instability in Huntington's disease. Brain 147, 1784-1798 (2024).28. E. Sapp, A. Boudi, A. Iwanowicz, J. Belgrad, R. Miller, R. Robertson, D. O'Reilly, K. Yamada, Y. Deng, M. Joni, X. Li, K. Kegel-Gleason, A. Khvorova, A. Reiner, N. Aronin, M. DiFiglia, Mutant huntingtin exon 1 protein detected in mouse brain with neoepitope antibody: effects of CAG repeat expansion, MutS Homolog 3 silencing and aggregation. Brain Commun 7, fcaf314 (2025).29. J. Belgrad, A. Summers, C. Landles, J. R. Greene, S. Hildebrand, E. Knox, E. Sapp, N. Yamada, R. Furgal, R. Miller, G. F. Osborne, K. Chase, E. Luu, J. Freedman, B. Bramato, N. McHugh, V. Benoit, D. O'Reilly, P. Greer, G. P. Bates, T. F. Vogt, R. Lee, D. Howland, M. DiFiglia, N. Aronin, A. Khvorova, Blocking somatic repeat expansion and lowering huntingtin via RNA interference synergize to prevent Huntington's disease pathogenesis in mice. bioRxiv, (2025).086013-058944530. J. Ochaba, A. M. Monteys, J. G. O'Rourke, J. C. Reidling, J. S. Steffan, B. L.Davidson, L. M. Thompson, PIAS 1 Regulates Mutant Huntingtin Accumulation and Huntington's Disease- Associated Phenotypes In Vivo. Neuron 90, 507-520 (2016).31. R. Goold, M. Flower, D. H. Moss, C. Medway, A. Wood-Kaczmar, R. Andre. P.Farshim, G. P. Bates, P. Holmans, L. Jones, S. J. Tabrizi, FAN1 modifies Huntington's disease progression by stabilizing the expanded HTT CAG repeat. Hum Mol Genet 28. 650-661 (2019).32. Z. L. McLean, D. Gao, K. Correia. J. C. L. Roy, S. Shibata, I. N. Famum. Z.Valdepenas-Mellor, M. Kovalenko, M. Rapuru, E. Morini, J. Ruliera, T. Gillis, D. Lucente, B. P. Kleinstiver, J. M. Lee, M. E. MacDonald, V. C. Wheeler, R. Mouro Pinto, J. F. Gusella, Splice modulators target PMS1 to reduce somatic expansion of the Huntington's disease-associated CAG repeat. Nat Commun 15, 3182 (2024).33. B. H. Laabs, C. Klein, J. Pozojevic, A. Domingo, N. Bruggemann, K. Grutz, R. L. Rosales, R. D. Jamora, G. Saranza, C. C. E. Diesta, M. Wittig, S. Schaake, M. Dulovic-Mahlow, J. Quismundo, P. Otto, P. Acuna, C. Go, N. Sharma, T. Multhaupt-Buell, U. Muller, H. Hanssen, F. Kilpert, A. Franke, A. Rolfs, P. Bauer, V. Dobricic, K. Lohmann, L. J. Ozelius, F. J. Kaiser, I. R. Konig, A. Westenberger. Identifying genetic modifiers of age-associated penetrance in X-linked dystonia-parkinsonism. Nat Commun 12, 3216 (2021).34. L. N. Campion, A. Mejia Maza, R. Yadav, E. B. Penney. M. G. Murcar, K. Correia, T. Gillis, C. Femandez-Cerado, M. S. Velasco-Andrada, G. P. Legarda, N. G. Ganza-Bautista, J. B. B. Lagarde, P J. Acuna. T. Multhaupt-Buell, G. Aldykiewicz, M. L. Supnet, J. K. De Guzman, C. Go, N. Sharma, E. L. Munoz, M. C. Ang, C. C. E. Diesta, D. C. Bragg, L. J. Ozelius, V. C. Wheeler, Tissue-specific and repeat length-dependent somatic instability of the X-linked dystonia parkinsonism-associated CCCTCT repeat. Acta Neuropathol Commun 10, 49 (2022). ”35. F. Morales, M. Vasquez, C. Santamaria, P. Cuenca, E. Corrales, D. G. Monckton, A polymorphism in the MSH3 mismatch repair gene is associated with the levels of somatic instability of the expanded CTG repeat in the blood DNA of myotonic dystrophy type 1 patients. DNA Repair (Amst) 40, 57-66 (2016).36. M. Flower, V. Lomeikaite, M. Ciosi, S. Cumming. F. Morales, K. Lo, D. Hensman Moss, L. Jones, P. Holmans, T.-H. Investigators, O. Consortium, D. G. Monckton, S. J.Tabrizi, MSH3 modifies somatic instability and disease severity in Huntington's and myotonic dystrophy type 1. Brain 142, 1876-1886 (2019).37. R. L. Boudreau, B. L. Davidson, Generation of hairpin-based RNAi vectors for biological and therapeutic application. Methods Enzymol 507, 275-296 (2012).38. E. Ayuso, F. Mingozzi, J. Montane, X. Leon, X. M. Anguela, V. Haurigot, S. A.Edmonson. L. Africa, S. Zhou. K. A. High, F. Bosch, J. F. Wright, High AAV vector purity results in serotype- and tissue-independent enhancement of transduction efficiency. Gene Ther 17, 503-510 (2010).39. V. C. Wheeler, W. Auerbach, J. K. White, J. Srinidhi, A. Auerbach, A. Ryan, M. P. Duyao, V. Vrbanac, M. Weaver, J. F. Gusella, A. L. Joyner, M. E. MacDonald, Lengthdependent gametic CAG repeat instability in the Huntington's disease knock-in mouse. Hum Mol Genet 8, 115-122 (1999).086013-058944540. S. Lee, B. Weisburd, J. Lee, K. Correia, S. Zeng, S. S. Park, J. W. Shin, D. E. Choi, K. H. Kim, J. H. Jang, T. Gillis, H. L. Rehm, J. F. Gusella, M. E. MacDonald, J. M. Lee, Significant underascertainment in Huntington's disease. Brain Commun 7, 194 (2025).41. R. L. Boudreau, R. M. Spengler. B. L. Davidson, Rational design of therapeutic siRNAs: minimizing off-targeting potential to improve the safety of RNAi therapy for Huntington's disease. Mol Ther 19, 2169-2177 (2011).42. J. L. McBride, M. R. Pitzer, R. L. Boudreau, B. Dufour, T. Hobbs, S. R. Ojeda, B. L. Davidson, Preclinical safety of RNAi -mediated HTT suppression in the rhesus macaque as a potential therapy for Huntington's disease. Mol Ther 19, 2152-2162 (2011).43. J. Miniarikova, I. Zanella, A. Huseinovic, T. van der Zon, E. Hanemaaijer, R. Martier, A. Koomneef, A. L. Southwell, M. R. Hayden, S. J. van Deventer, H. Petry, P. Konstantinova, Design, Characterization, and Lead Selection of Therapeutic miRNAs Targeting Huntingtin for Development of Gene Therapy for Huntington's Disease. Mol Ther Nucleic Acids 5, e297 (2016).

Claims

WHAT IS CLAIMED IS:

1. A method for modeling or projecting therapeutic effect on somatic instability and / or cell survival in Huntington’s Disease, comprising:encoding a plurality' of cell objects with cell attributes and cell functions to represent medium spiny neurons (MSN), wherein:the cell attributes compnse cell age, cytosine-adenine-guanine (CAG) repeat length, and an indication of treated or untreated treatment status; and the cell functions are configured to facilitate simulated modification of individual cell objects;instantiating at least one cell collection of the cell objects, the cell collection having a defined CAG repeat length, the cell collection being encoded with population functions configured to facilitate:a simulated cell lifetime of somatic expansion events for the CAG repeat length across the cell objects in a cell collection, andapplying a therapeutic to the cell objects in the cell collection using a rate of cell transduction;simulating, using the population functions and the cell functions, and based on the cell attributes and a defined CAG repeat expansion rate, lifecycles for the cell objects in the cell collection, wherein:the at least one cell collection is configured to simulate Huntington’s Disease progression under therapeutic treatment;the simulation comprises a life cycle of mutation events at the defined CAG repeat expansion rate across the cell objects in the cell collection; and mutation rates for cell objects receiving said therapeutic treatment are modified to reflect reduced CAG repeat expansion; anddetermining the therapeutic effect on somatic instability and / or cell survival in Huntington’s Disease based on the simulating, thereby modeling or projecting therapeutic effect on somatic instability and / or cell survival in Huntington’s Disease.

2. The method of claim 1, further comprising instantiating two or more cell collections, each having defined CAG repeat lengths, with at least one cell collection configured to simulate Huntington’s Disease progression without therapeutic treatment, such that the at least one cell collection is configured to function as a control.

3. The method of claim 2, wherein determining the therapeutic effect on somatic instability and / or cell survival in Huntington’s Disease based on the simulating comprises:identifying cell objects as healthy if their CAG repeat length remains below a threshold after the simulating, and comparing a quantity of healthy cell objects in the at least one cell collection configured to simulate Huntington’s Disease progression without therapeutic treatment to a quantity of healthy cell objects in the cell collection configured to simulate Huntington’s Disease progression under therapeutic treatment, at various time intervals during the lifecycles of the cell objects.

4. The method of claim 3, wherein the threshold is above 50, 80, 100, 150, 200 CAG repeats, or between about 100-150 or about 100-200 CAG repeats.

5. The method of claim 2, wherein determining the therapeutic effect on somatic instability and / or cell survival in Huntington's Disease based on the simulating further comprises determining a therapeutic benefit for a treated population of cell objects as a temporal delay in reaching a predefined Huntington’s Disease milestone at one or more of the various time intervals compared to an untreated population of cell objects.

6. The method of claim 5, wherein predefined disease milestones comprise onset of volumetric changes to caudate and putamen, optionally as determined by one or more of magnetic resonance imaging (MRI), elevated cerebrospinal fluid (CSF) neurofilament like protein (NFL), motor symptoms, and / or Huntington’s Disease stages HD ISS 1, 2, and / or 3.

7. The method of claim 5, wherein determining the therapeutic effect on somatic instability and / or cell survival in Huntington’s Disease based on the simulating further comprises determining CAG repeat length for each cell object in each population at time intervals, and comparing CAG repeat lengths from the treated population of cell objects to CAG repeat lengths from the untreated population of cell objects at the time intervals.

8. The method of claim 5, wherein the therapeutic applied is a gene therapy.

9. The method of claim 5, wherein the therapeutic applied is a siRNA targeting MSH3, miRNA targeting MSH3, anti-sense oligonucleotide targeting MSH3, zinc finger protein tosilence MSH3 expression, CRISPR-based means to reduce or silence MSH3 expression, or an epigenetic effector to reduce or silence MSH3 expression.

10. The method of claim 1 , further comprising repeating the instantiating and the simulating for populations of cell objects with different CAG repeat lengths and ages of therapeutic interv ention, and determining the therapeutic effect on somatic instability and / or cell survival in Huntington’s Disease based on CAG repeat length and therapeutic intervention age combination.

11. The method of claim 10, wherein repeating the instantiating and the simulating for populations of cell objects with different CAG repeat lengths and cell ages, and determining the therapeutic effect on somatic instability and / or cell survival in Huntington’s Disease based on CAG repeat length and age at therapeutic intervention comprises defining a model parameter space comprising a range of CAG repeat lengths and a range of cell ages, and executing simulations across the parameter space to generate predicted therapeutic benefits based on CAG repeat length and age.

12. The method of claim 10, further comprising identifying or determining a treatable individual or population of individuals based on their ages and CAG repeat lengths, and optionally determining if said individual or population of individuals is eligible for treatment based upon criteria such as age, CAG repeat lengths, predicted therapeutic effect and / or costs.

13. The method of claim 12, wherein the treatable population of individuals is determined for a clinical application or clinical trial of the therapeutic.

14. The method of claim 10, further comprising generating a map visualization of therapeutic effect values as a function of CAG repeat length and cell age when the therapeutic was applied.

15. The method of claim 10, further comprising repeating the instantiating and the simulating for cell collections of cell objects with CAG repeat lengths from 36-60 repeating units and cell ages from about 5-100 years.

16. The method of claim 1, wherein the CAG repeat length expansion rate is user defined.

17. The method of claim 16, wherein the CAG repeat length expansion rate changes based on the CAG repeat length.

18. The method of claim 17, wherein the CAG repeat length expansion rate is associated with disease progression milestones at known CAG repeat lengths and age combinations.

19. The method of claim 1, wherein the cell functions comprise obtaining CAG repeat lengths, determining a CAG repeat length for each cell object, triggering modeled therapeutic transduction events, obtaining therapeutic transduction status, modelling mutating a CAG repeat, and modelling mutating a CAG repeat in context of a cell to which a therapy has been administered.

20. The method of claim 1, wherein the population functions comprise modelling mutating all cells in a population, modelling administering a gene therapy, modelling mutating cells after therapy administration, running a lifecycle modelling algorithm for all cells in a population, and running a lifecycle modelling algorithm with modeled therapy.

21. The method of claim 1, wherein cell ages, CAG repeat lengths, treatment statuses, population sizes, characteristics of the therapeutic, rate of transduction, and / or the MSN somatic expansion rate comprise user defined variables.

22. The method of claim 1, wherein the plurality of cell objects comprises up to about 100, 1000, 1000000, or 10000000 cell objects.

23. The method of claim 1, wherein the cell collection comprises 1 or 2, up to about 100, 1000, 1000000, or 10000000 cells.

24. The method of claim 1, wherein the cell collection represents a single human patient.

25. A method for modeling or projecting therapeutic effect on somatic instability and cell survival in repeat expansion disorders, comprising:encoding a plurality of cell objects with cell attributes and cell functions to represent human cells, wherein:the cell attributes comprise cell age, a repeat length, and an indication of treated or untreated treatment status; andthe cell functions are configured to facilitate simulated modification of individual cell objects;instantiating a cell collection of the cell objects, the cell collection having a defined repeat length, the cell collection being encoded with population functions configured to facilitate:a simulated cell lifetime of somatic expansion events for repeat lengths across the cell objects in a cell collection; andapplying a therapeutic to the cell obj ects in the cell collection using a rate of cell transduction;simulating, using the population functions and the cell functions, and based on the cell attributes and a defined somatic expansion rate, lifecycles for the cell objects in the cell collection, wherein:the cell collection is configured to simulate disease progression under therapeutic treatment;the simulation comprises a life cycle of mutation events at the defined expansion rate across the cell objects in the cell collection; and mutation rates for cell objects receiving simulated therapeutic treatment are modified to reflect reduced somatic repeat expansion; and determining the therapeutic effect on somatic instability and cell survival based on the simulating, thereby modeling or projecting therapeutic effect on somatic instability' and cell survival in repeat expansion disorders.

26. The method of claim 25, further comprising instantiating two or more cell collections, each having defined repeat lengths, with at least one cell collection configured to simulate natural disease progression without therapeutic treatment, such that the at least one cell collection is configured to function as a control.

27. The method of claim 25, wherein determining the therapeutic effect on somatic instability and cell survival based on the simulating comprises:identifying cell objects as healthy if their repeat length remains below a threshold after the simulating, and comparing a quantity of healthy cell objects in the at least one cell collection configured to simulate natural disease progression without therapeutic treatment toa quantity of healthy cell objects in the cell collection configured to simulate disease progression under therapeutic treatment, at various time intervals during the lifecycles of the cell objects.

28. The method of claim 27, wherein determining the therapeutic effect on somatic instability7and cell survival based on the simulating further comprises determining a therapeutic benefit for a treated population of cell objects as a temporal delay in reaching a predefined disease milestone at one or more of the various time intervals compared to an untreated population of cell objects.

29. The method of claim 26. wherein determining the therapeutic effect on somatic instability and cell survival based on the simulating further comprises determining the repeat length for each cell object in each cell collection at time intervals, and comparing repeat lengths from the treated population of cell objects to repeat lengths from the untreated population of cell objects at the time intervals.

30. The method of claim 25, wherein the therapeutic applied is a gene therapy.

31. The method of claim 25, further comprising repeating the instantiating and the simulating for populations of cell objects with different repeat lengths and ages of therapeutic intervention, and determining the therapeutic effect on somatic instability7and cell survival based on repeat length and therapeutic intervention age combination.

32. The method of claim 29, further comprising determining a treatable population of individuals for a clinical trial of the therapeutic based on their ages and repeat lengths.

33. The method of claim 25, further comprising repeating the instantiating and the simulating for cell collections of cell objects with repeat lengths from 36-60 repeating units and cell ages from about 5-100 years.

34. The method of claim 25, wherein the somatic expansion rate changes based on repeat length.

35. The method of claim 25, wherein the plurality of cell objects comprises up to 100, 1000, 1000000. or 10000000 cell objects.

36. The method of claim 25, wherein the cell collection comprises 1 or 2 or up to 100, 1000, 1000000, or 10000000 cells.

37. The method of claim 25, wherein the cell collection represents a single human patient.

38. The method of claim 25, wherein the repeat expansion disorder is a genetic neurodegenerative disorder.

39. The method of claim 38, wherein the repeat expansion disorder is Huntington’s disease (HTT. CAG), Huntington’s disease-like 2 (JPH3, CTG / CAG), spinal and bulbar muscular atrophy or Kennedy disease (AR, CAG), dentatorubral-pallidoluysian atrophy (ATN1, CAG), spinocerebellar ataxia (SCA) type 1 (ATXN1, CAG), SCA type 2 (ATXN2, CAG), SCA type 3 / Machado-Joseph disease (ATXN3. CAG), SCA type 6 (CACNA1A, CAG), SCA type 7 (ATXN7, CAG), SCA type 8 (SCA8, CTG / CAG), SCA type 10 (SCA10, ATTCT / ATTTT), SCA type 12 (SCA12, CAG / CTG), SCA type 17 (SCA17, CAG / CAA), SCA type 27B (FGF14, GAA), SCA type 31 (SCA31, TGGAA), SCA type 36 (SCA36, GGCCTG), SCA type 37 (SCA37. ATTTC / ATTTT), SCA type 51 (SCA51). myotonic dystrophy type 1 (DMPK, CTG), myotonic dystrophy type 2 (CA77P / ZNF9, CCTG), Fragile X syndrome (FMRI, CGG), fragile X-associated tremor / ataxia syndrome (FXTAS, CGG premutation), fragile X-associated primary ovarian insufficiency (FXPOI, CGG premutation), C9orf72-associated amyotrophic lateral sclerosis and frontotemporal dementia C9orf72, GGGGCC), neuronal intranuclear inclusion disease (NOTCH2NLC, CGG), oculopharyngeal myopathy and leukoencephalopathy (OPML, CGG), oculopharyngodistal myopathy types 1-3 (OPDM1-3, CGG), cerebellar ataxia, neuropathy, and vestibular areflexia syndrome (CANVAS) (RFC1, AAGGG), familial adult myoclonic epilepsy 1 (FAME 1), familial adult myoclonic epilepsy 2 (FAME2), familial adult myoclonic epilepsy 3 (FAME3), familial adult myoclonic epilepsy 4 (FAME4), familial adult myoclonic epilepsy (FAME6), familial adult myoclonic epilepsy 7 (FAME7), wherein any of the FAME is caused by pentanucleotide repeats such as TTTCA, TTTTA, ATTTC, and ATTTT in intronic regions, Fuchs endothelial comeal dystrophy (TCF4, CTG / CAG intronic repeat), autosomal dominant tubulointerstitial kidney disease (MUC1, mono-C frameshift repeat), X-linked intellectual developmentaldisorders / Partington syndrome (AR , GCG / CGC repeats); panhypopituitarism with grow th hormone deficiency (SOX3, GCN), holoprosencephaly type 5 (ZIC2, GCN); VACTERL association (ZIC3. GCC), hereditary sensory and autonomic neuropathy type VIII (PRDM12. GCC), Fragile XE syndrome (AFF2, CCG), a small-repeat disorder in PRNP, or a smallrepeat disorder in MARCH F6.

40. A non-transitory computer readable medium having instructions thereon, the instructions, when executed by a computer, causing the computer to perform operations for modeling or projecting therapeutic effect on somatic instability and cell survival in repeat expansion disorders, the operations comprising:encoding a plurality of cell objects with cell attributes and cell functions to represent human cells, wherein:the cell attributes comprise cell age, a repeat length, and an indication of treated or untreated treatment status; andthe cell functions are configured to facilitate simulated modification of individual cell objects;instantiating a cell collection of the cell objects, the cell collection having a defined repeat length, the cell collection being encoded with population functions configured to facilitate:a simulated cell lifetime of somatic expansion events for repeat lengths across the cell objects in a cell collection; andapplying a therapeutic to the cell obj ects in the cell collection using a rate of cell transduction;simulating, using the population functions and the cell functions, and based on the cell attributes and a defined somatic expansion rate, lifecycles for the cell objects in the cell collection, wherein:the cell collection is configured to simulate disease progression under therapeutic treatment;the simulation comprises a life cycle of mutation events at the defined expansion rate across the cell objects in the cell collection; and mutation rates for cell objects receiving simulated therapeutic treatment are modified to reflect reduced somatic repeat expansion; anddetermining the therapeutic effect on somatic instability and cell survival based on the simulating, thereby modeling or projecting therapeutic effect on somatic instability and cell survival in repeat expansion disorders.

41. The medium of claim 40, further comprising instantiating two or more cell collections, each having defined repeat lengths, with at least one cell collection configured to simulate natural disease progression without therapeutic treatment, such that the at least one cell collection is configured to function as a control.

42. The medium of claim 40, wherein determining the therapeutic effect on somatic instability and cell survival based on the simulating comprises:identifying cell objects as healthy if their repeat length remains below a threshold after the simulating, and comparing a quantify of healthy cell objects in the at least one population configured to simulate natural disease progression without therapeutic treatment to a quantity of healthy cell objects in the cell collection configured to simulate disease progression under therapeutic treatment, at various time intervals during the lifecycles of the cell objects.

43. The medium of claim 40, wherein determining the therapeutic effect on somatic instability and cell survival based on the simulating further comprises determining a therapeutic benefit for a treated population of cell objects as a temporal delay in reaching a predefined disease milestone at one or more of the various time intervals compared to an untreated population of cell objects.

44. The medium of claim 40, wherein determining the therapeutic effect on somatic instability and cell survival based on the simulating further comprises determining the repeat length for each cell object in each cell collection at time intervals, and comparing repeat lengths from the treated population of cell objects to repeat lengths from the untreated population of cell objects at the time intervals.

45. The medium of claim 40, further comprising repeating the instantiating and the simulating for populations of cell objects with different repeat lengths and ages of therapeutic intervention, and determining the therapeutic effect on somatic instability and cell survival based on repeat length and therapeutic intervention age combination.

46. The medium of claim 45, further comprising identifying or determining a treatable individual or population of individuals for treatment or a clinical trial of the therapeutic based on their ages and repeat lengths, and optionally determining if said individual or population of individuals is eligible for treatment based upon criteria such as age, repeat length, predicted therapeutic effect and / or costs.

47. The medium of claim 45, further comprising repeating the instantiating and the simulating for cell collections of cell objects with repeat lengths from 36-60 repeating units and cell ages from about 5-100 years.

48. The medium of claim 40, wherein the somatic expansion rate changes based on repeat length.

49. The medium of claim 40, wherein the plurality of cell objects comprises up to 100, 1000, 1000000, or 10000000 cell objects.

50. The medium of claim 40, wherein the cell collection comprises up to 100, 1000, 1000000, or 10000000 cells.

51. The medium of claim 40, wherein the cell collection represents a single human patient.

52. The method of claim 11, wherein the model parameter space further comprises:(i) a treated-fraction parameter representing a proportion of the cell objects representing medium spiny neurons (MSNs) that receive simulated therapeutic treatment via the rate of cell transduction, and(ii) a knockdown parameter representing a modeled reduction in expression of a target gene product in the cell objects receiving simulated therapeutic treatment,and wherein executing simulations across the parameter space comprises varying the treated-fraction parameter and the knockdown parameter.

53. The method of claim 52, further comprising determining, based on the executing simulations across the parameter space, a minimum treated-fraction value and a minimum knockdown value sufficient to produce a predicted therapeutic benefit comprising a temporaldelay of greater than 5 years in reaching a predefined Huntington’s Disease milestone relative to an untreated population.

54. The method of claim 53, wherein the minimum treated-fraction value is at least 50% and the minimum knockdow n value is at least 50%.

55. The method of claim 53, further comprising outputting the minimum treated-fraction value and the minimum knockdown value as a minimum target product profile for the therapeutic.

56. The method of claim 52. wherein the target gene product comprises MSH3.

57. The method of claim 1, further comprising evaluating a candidate therapeutic configuration by:(i) simulating, for the candidate therapeutic configuration, a treated-fraction value for the MSNs and a knockdown value for the target gene product, and(ii) determining whether the candidate therapeutic configuration is predicted to produce a therapeutic benefit comprising a temporal delay of greater than 5 years in reaching a predefined Huntington’s Disease milestone relative to an untreated population.

58. The method of claim 57, wherein determining whether the candidate therapeutic configuration is predicted to produce the therapeutic benefit comprises determining that: (i) the treated-fraction value is at least 50%, and(ii) the knockdown value is at least 50%.

59. The method of claim 57, further comprising selecting the candidate therapeutic configuration for further development or a clinical application when the candidate therapeutic configuration is predicted to produce the therapeutic benefit, and rejecting the candidate therapeutic configuration otherwise.

60. The method of claim 57, further comprising, when the candidate therapeutic configuration is rejected, modifying at least one of (i) the rate of cell transduction or (ii) the knockdown value, and repeating the simulating and the evaluating.

61. The medium of claim 40, wherein the repeat expansion disorder is a genetic neurodegenerative disorder.

62. The medium of claim 61, wherein the repeat expansion disorder is Huntington’s disease (HTT. CAG), Huntington’s disease-like 2 (JPH3, CTG / CAG), spinal and bulbar muscular atrophy or Kennedy disease (AR, CAG). dentatorubral-pallidoluysian atrophy (ATN1, CAG), spinocerebellar ataxia (SCA) type 1 (ATXN1, CAG), SCA type 2 (ATXN2, CAG), SCA type 3 / Machado-Joseph disease (ATXN3, CAG), SCA type 6 (CACNA1A, CAG), SCA type 7 (A7A7V7, CAG), SCA type 8 (SCA8, CTG / CAG), SCA type 10 (SCAIO, ATTCT / ATTTT), SCA type 12 (SCA12, CAG / CTG), SCA type 17 (SCA17, CAG / CAA), SCA type 27B (FGF14, GAA), SCA type 31 (SCA31, TGGAA). SCA type 36 (SCA36, GGCCTG), SCA type 37 (SCA37, ATTTC / ATTTT), SCA type 51 (SCA51), myotonic dystrophy type 1 (DMPK, CTG), myotonic dystrophy type 2 (CAB / ZNF9, CCTG), Fragile X syndrome (FMRI, CGG), fragile X-associated tremor / ataxia syndrome (FXTAS, CGG premutation), fragile X-associated primary ovarian insufficiency (FXPOI, CGG premutation), C9orf72-associated amyotrophic lateral sclerosis and frontotemporal dementia (C9orf72, GGGGCC), neuronal intranuclear inclusion disease (NOTCH2NLC, CGG), oculopharyngeal myopathy and leukoencephalopathy (OPML, CGG), oculopharyngodistal myopathy types 1-3 (OPDM1-3, CGG), cerebellar ataxia, neuropathy , and vestibular areflexia syndrome (CANVAS) (RFC1, AAGGG), familial adult myoclonic epilepsy 1 (FAME 1). familial adult myoclonic epilepsy 2 (FAME2), familial adult myoclonic epilepsy 3 (FAME3), familial adult myoclonic epilepsy 4 (FAME4), familial adult myoclonic epilepsy (FAME6), familial adult myoclonic epilepsy 7 (FAME7), wherein any of the FAME is caused by pentanucleotide repeats such as TTTCA, TTTTA, ATTTC, and ATTTT in intronic regions. Fuchs endothelial comeal dystrophy (TCF4, CTG / CAG intronic repeat), autosomal dominant tubulointerstitial kidney disease (MUC1, mono-C frameshift repeat), X-linked intellectual developmental disorders / Partington syndrome (ARX, GCG / CGC repeats); panhypopituitarism with grow th hormone deficiency (SOX3, GCN), holoprosencephaly type 5 (ZIC2, GCN); VACTERL association (ZIC3, GCC), hereditary sensors’ and autonomic neuropathy type VIII (PRDM12, GCC), Fragile XE syndrome (AFF2, CCG), a small-repeat disorder in PRNP, or a smallrepeat disorder in MARC HF6.

Images