Mountain peaks at sunrise
A welcoming home for HD research

Helping people make sense of Huntington's disease.

An open AI research home for people touched by Huntington's disease. Chat with the actual papers, in plain language, in any of 22 languages.

Start by telling us who you are arrow_downward

We are data scientists, not doctors. This is a research and education tool. For care decisions, please reach out to HDSA, HDBuzz, HDYO, or your own neurologist.

Who are you?

Tell us once. The hub adapts.

Pick the lens that matches you. The chat, the explainers, and the suggestions on this page change to fit.

Autonomous research agents
Refreshed today at 11:04 UTC · 20 papers · 30 trials · 20 targets
cloud_sync data_fetcher
Daily data refresh landed on main
today at 11:04 UTC
draft preprint_scout
Preprint scout indexed 10 HD preprints from biorxiv + medrxiv
today at 11:04 UTC
science experiment_runner
Experiment #005 published
May 5
biotech hypothesis_refiner
Hypothesis refiner reviewed 5 drug ideas
May 5
menu_book paper_scout
Paper scout ingested 76 papers
May 5
graph_4 autoresearch
Full pipeline run · 1/1 agents ok
May 5
experiment paper_analyzer
Paper analyzer scored 10 abstracts
May 5
Daily refresh · runningEdge pipeline · runs weeklyExperiments shipped · 5Active hypotheses · 5
What we've learned

Five experiments. Honest writeups. Reproducible.

Each one asks one question, runs Gemma 4 or its predecessors over real PubMed papers, and ships the result, including what didn't work.

EXP-001 Mar 27, 2026

First-look LLM triage: 22 HD abstracts on Llama 3.1

Post-translational modifications of huntingtin (phosphorylation, SUMOylation, O-GlcNAcylation) are getting significant research attention as potential drug targets.

Model: llama3.1:8b Read the writeup arrow_forward
EXP-002 Mar 30, 2026

Full-text deep read: 16 HD papers on Llama 3.1

The K845N variant of LIG1 confers enhanced substrate discrimination and increased repair fidelity, suppressing somatic CAG expansion in mice.

Model: llama3.1:8b Read the writeup arrow_forward
EXP-003 Apr 02, 2026

Model comparison: qwen3.5 vs Llama 3.1 on the EXP-002 corpus

A model-comparison run: re-analyzed the Experiment #2 corpus with qwen3.5:27b. Writeup available with the raw results and reproducibility notes.

Model: qwen3.5:27b Read the writeup arrow_forward
EXP-003-GEMMA Apr 02, 2026

Gemma 4 finds what Llama missed: multi-mechanism HD pathology

Neurodegeneration is not attributable to a single mechanism but results from a complex, interacting pathogenic network involving protein misfolding/aggregation, mitochondrial dysfunction, neuroinflammation, metabolic dysregulation, and gut dysbiosis, necess...

Model: gemma4:latest Read on GitHub open_in_new
EXP-004 Apr 07, 2026

Somatic CAG expansion drug screen (75-paper sweep)

Screened 75 papers across MSH3, FAN1, PMS1, MLH1, LIG1, the MutSbeta to MutLgamma to LIG1 pathway now driving HD drug development at LoQus23, Harness, Skyhawk, and Rgenta.

Model: gemma4:latest Read the writeup arrow_forward
EXP-005 Apr 19, 2026

Expanded corpus: the two-track HD pathology model

"Two-Track" Pathology Model: HD is increasingly seen as a confluence of two distinct processes: Track 1 (primary genetic/structural damage to HTT) and Track 2 (secondary metabolic and immune failure).

Model: gemma4:latest Read on GitHub open_in_new
How the Hub works

Pick a topic. Read what it means. Then do something with it.

Huntingtin-lowering (ASOs and gene therapy)

A strategy worth watching A simple idea, big stakes

Picture the huntingtin protein as a faulty thermostat that someone left turned up too high. Huntingtin-lowering programs try to turn the dial back down, so the brain has less of the harmful version to deal with in the first place.

What it is

A family of treatments that tries to make the brain produce less of the huntingtin protein. Some are repeat injections, some are a one-time gene therapy.

Read this at:
Do something with it
What the Hub is tracking today
20
Recent Papers
30
Active Trials
21
Recruiting Now
16+
Known Targets

Updated May 19, 2026 at 11:04 AM. Auto-refreshed daily from PubMed, ClinicalTrials.gov, HDBuzz, and Open Targets.

What's Happening Now

HD News + Breakthroughs

More on HDBuzz open_in_new
Fresh from PubMed

Latest Research Papers

All HD papers open_in_new
2026 May — CNS & neurological disorders drug target
protein structure

Understanding Huntington's Disease: Epidemiology, Mechanisms, and Modeling Approaches.

Huntington's disease (HD) is a monogenic, autosomal dominant neurodegenerative disorder. Huntington's disease is caused by a CAG trinucleotide repeat expansion in exon 1 of the huntingtin gene, which

PubMed open_in_new
2026 Apr — Molecular therapy : the journal of the A
gene therapyprotein structure

Treatment of Huntington's disease with a pan-HTT-targeting CRISPR nuclease.

Huntington's disease (HD) is an inherited neurodegenerative disorder caused by an expansion of a CAG trinucleotide repeat in the huntingtin (HTT) gene, which leads to a mutant protein that destroys ne

PubMed open_in_new
2026 Jun — Disease models & mechanisms
protein structure

Induced pluripotent stem cells from a transgenic minipig model of Huntington's disease reveal early

Huntington's disease (HD) is a neurodegenerative autosomal dominant hereditary disease caused by a CAG triplet repeat expansion mutation in the gene encoding the huntingtin (HTT) protein. The main fea

PubMed open_in_new
2026 May — Laeknabladid
autophagyprotein structure

[Huntington´s disease - overview].

Huntington's disease (HD) is a rare, autosomal dominant neurodegenerative disorder characterized by progressive movement disturbances, psychiatric symptoms, and cognitive impairment. The disease is ca

PubMed open_in_new
2026 Jun — Neurobiology of disease
gene therapyprotein structure

Propagation of gut-injected AAV2/1-Htt171-82Q to the brain induces Huntington's disease-like patholo

Huntington's disease (HD) is a progressive neurodegenerative disorder that leads to thbe production of pathogenic mutant huntingtin (mHtt) protein and is characterized by motor and psychiatric symptom

PubMed open_in_new
2026 Apr — Journal of Huntington's disease
somatic expansion

Towards AI-driven prediction of HTT CAG size in super-expanded human spiny projection neurons from H

Somatic instability (SI) of the CAG tract in HTT is a major driver of neurodegeneration of Spiny Projection Neurons (SPNs), the primary neuronal subtype affected in Huntington's disease (HD). SPNs can

PubMed open_in_new
Reasons for Optimism

Active Clinical Trials

30 trials, 163,742 patients enrolled, 21 recruiting now

ClinicalTrials.gov open_in_new
NCT ID Title Sponsor Phase Status Intervention
NCT06807892 Comparison of Two Cognitive-Motor Rehabilitation A University Hospital, Ange NA Recruiting Synergistic management; Sequential manag
NCT05686551 GENERATION HD2. A Study to Evaluate the Safety, Bi Hoffmann-La Roche PHASE2 Active Not Recruiting Tominersen; Placebo
NCT06203106 NYSCF Scientific Discovery Biobank New York Stem Cell Founda N/A Recruiting Biological Sample Collection
NCT06941662 Huntington's Disease Biobank: Advancing Remote Mon Stanford University N/A Recruiting computer vision, remote monitoring, bloo
NCT05402488 Exploring the Recovery Function of Sleep in Neurod University of Zurich NA Recruiting Auditory stimulation
NCT06254482 An Extension Study to Evaluate the Long-Term Safet Novartis Pharmaceuticals PHASE2 Active Not Recruiting Votoplam
NCT02252380 ExAblate Transcranial MRgFUS for the Management of InSightec NA Active Not Recruiting Transcranial ExAblate System
NCT06209515 Sociodemographic Factors and Criminal Behaviour Pr Kuopio University Hospita N/A Active Not Recruiting
NCT03233646 Retinal Imaging in Neurodegenerative Disease Duke University N/A Recruiting Retinal and Choroidal Imaging
NCT05243017 Safety and Efficacy of AMT-130 in European Adults UniQure Biopharma B.V. PHASE1, PHASE2 Active Not Recruiting intra-striatal rAAV5-miHTT
Open Research

Published Research Ideas

AI-assisted repurposing ideas with explicit uncertainty. These are starting points for review, not recommendations or validated findings.

Read Experiment #1: Full Analysis arrow_forward
Tocilizumab Exploring
Target: IL-6 / neuroinflammation

IL-6 receptor blockade may reduce inflammation from mutant huntingtin aggregation

Latest score: 80/100 Support: 0 Concerns: 0 Updated: 2026-03-29
Metformin Exploring
Target: mTOR / AMPK / autophagy

Promotes autophagy which may help clear mutant HTT aggregates

Latest score: 72/100 Support: 0 Concerns: 0 Updated: 2026-03-29
Rapamycin Exploring
Target: mTOR

mTOR inhibitor demonstrated clearance of mutant HTT in preclinical models

Latest score: 68/100 Support: 0 Concerns: 0 Updated: 2026-03-29
Riluzole Known/Tested
Target: Glutamate excitotoxicity

Glutamate release inhibitor, may reduce excitotoxicity in HD

Latest score: 60/100 Support: 0 Concerns: 1 Updated: 2026-03-29
Lithium Exploring
Target: TDP-43 / GSK-3β

May reduce TDP-43 phosphorylation and aggregation via GSK-3β inhibition, relevant to newly discovered HD-TDP-43 connection

Latest score: 55/100 Support: 0 Concerns: 0 Updated: 2026-03-29

Important: These hypotheses are triage artifacts, not evidence of efficacy. They have not been clinically validated, experimentally confirmed, or expert-reviewed unless explicitly stated.

These cards are generated from data/hypotheses_tracker.json, not hand-picked homepage copy. Are you an HD researcher? We'd love your feedback. Review and discuss on GitHub

waving_hand

Who This Is For

This is best viewed as open research infrastructure: a place where data scientists, AI engineers, bioinformaticians, and researchers can test whether agent workflows actually help with literature review, hypothesis triage, and research communication.

Data Scientists + AI/ML Engineers

Run our autonomous research agents. Fork the repo. Try different models. Add experiments. This is open-source infrastructure you can build on.

Researchers + Domain Experts

Review our AI-generated hypotheses. Spot what's promising and what's wrong. Your expertise is what turns computational ideas into real science. Join the discussion.

Students + Curious Builders

Want to learn how AI applies to drug discovery? Read our experiment reports — we show every step, what worked, and what didn't. No PhD required to follow along.

For patients and families: This is not a medical resource. For support, guidance, and verified medical information, please go to HDSA, HDBuzz, or HDYO. They have doctors and scientists reviewing every word. We don't.

Trusted Organizations

Resources for Patients, Families + Researchers

These organizations are the real experts. We link to them because they do essential work.

Make a Difference

Ways to Get Involved

Whether you're a patient, family member, researcher, developer, or just someone who cares.

volunteer_activism

Join a Clinical Trial

21 HD trials are actively recruiting right now. Your participation directly accelerates the path to treatment.

Find recruiting trials open_in_new
code

Contribute to Open Source

Our research agent, literature scanner, and drug repurposing tools are all open source. Data scientists, ML engineers, and bioinformaticians welcome.

View on GitHub open_in_new
record_voice_over

Share Your Story

Patient and family stories drive awareness and funding. Share your experience with HDSA or on social media to help others feel less alone.

HDSA Get Involved open_in_new
favorite

Donate

Research funding saves lives. HDSA and CHDI Foundation fund the science that makes breakthroughs like AMT-130 possible.

Donate to HDSA open_in_new
forum

Share Feedback or Ideas

Are you an HD researcher? See something interesting in our hypotheses? Spot an error? We want to hear from you — every correction makes this better.

Join the Discussion open_in_new