A single strand of DNA measures about two nanometers across. That's roughly fifty thousand times narrower than this period at the end of this sentence. Packed inside that impossibly tiny thread is every instruction your body has ever followed - including, it turns out, some very specific notes about whether your brain might one day start dismantling itself. A sweeping new review in the Journal of Clinical Investigation just laid out exactly how far we've come in reading those notes, and honestly? It's both thrilling and a little humbling.

Your Genome Has a Lot to Say (If You Know How to Listen)

Here's the thing. For decades, scientists studying neurodegenerative diseases like Alzheimer's, Parkinson's, and ALS were basically trying to read a novel with half the pages ripped out. Traditional genetic testing only looked at small, known chunks of DNA - the obvious suspects. But whole-genome sequencing changed the game entirely. Instead of checking under a few lampposts, researchers can now flood the entire street with light.

Grassano and colleagues at the NIH walk us through the tech revolution that made this possible [1]. Short-read sequencing gave us speed. Long-read sequencing (courtesy of companies like PacBio and Oxford Nanopore) gave us the ability to read continuous stretches spanning tens of thousands of base pairs - think of it as finally being able to read whole paragraphs instead of individual letters. And optical genome mapping? That lets scientists literally visualize the architecture of chromosomes, catching the big structural shake-ups that shorter reads completely miss.

The Genetic Villains We Didn't See Coming

Plot twist: some of the most devastating genetic culprits behind neurodegeneration aren't simple typos in your DNA. They're structural catastrophes - entire chunks of genetic material duplicated, deleted, flipped backwards, or stuttering like a broken record.

Take the C9orf72 gene. A six-letter sequence (GGGGCC) normally repeats a handful of times. But in people with familial ALS and frontotemporal dementia, that repeat can balloon to hundreds or even thousands of copies [2]. This single expansion is now recognized as the most common genetic cause of both ALS and FTD, accounting for a staggering 34% of familial ALS cases. Population studies in the review reveal that pathogenic-range repeat expansions affect roughly 1 in 283 people. That number should make you sit up straighter.

Copy number gains in SNCA cause autosomal dominant Parkinson's. Duplications in APP do the same for Alzheimer's. The MAPT gene carries a massive inversion that defines major tauopathy risk. None of this shows up on a standard genetic test from five years ago.

GWAS: The World's Most Expensive Game of "Spot the Difference"

Genome-wide association studies (GWAS) have been the workhorses of genetic discovery, scanning millions of DNA markers across thousands of people to find which tiny variations pop up more often in sick folks versus healthy ones [3]. For Alzheimer's alone, GWAS have flagged over 75 risk loci. But here's the catch - those 75 loci only explain about 15% of who actually gets the disease. The remaining 85%? Still lurking somewhere in the genetic dark matter, tangled up with environmental factors, lifestyle, and probably sheer cosmic luck.

Polygenic risk scores try to add up all those small nudges into one number, and people in the highest percentiles do show several-fold greater odds of developing Alzheimer's. But "several-fold greater odds" isn't the same as a diagnosis. It's more like a weather forecast that says "increased chance of rain" while you're deciding whether to carry an umbrella for the next forty years.

The Multi-Omics Power Move

Look. Reading DNA alone is like reading a recipe without checking what actually comes out of the oven. That's why the most exciting part of this review is the push toward multi-omics - combining genomics with transcriptomics (what genes are actually being turned on), proteomics (what proteins are being made), and functional validation (does this variant actually break something?).

RNA sequencing alone has bumped diagnostic success rates by 7% to over 30% by catching things like abnormal splicing and cryptic exon inclusions that DNA sequencing alone would miss [4]. Single-cell transcriptomics can now reveal which specific brain cell types are vulnerable in each disease - because it turns out neurodegeneration is less like a citywide blackout and more like someone systematically unscrewing particular lightbulbs.

Then there's the AI angle. Machine learning models like AlphaMissense are getting surprisingly good at predicting whether a given genetic variant is going to cause trouble, though the review is careful to note that human oversight isn't going anywhere soon. Your robot overlords aren't replacing genetic counselors just yet.

The Diversity Problem Nobody Wants to Talk About

One of the sharpest points in this review: the vast majority of GWAS data comes from people of European ancestry [5]. That means our polygenic risk scores, our variant databases, our whole framework for calling a genetic change "normal" or "pathogenic" - it's all built on a skewed foundation. Fixing this isn't just a fairness issue. It's a scientific one. You can't claim to understand human genetics by only studying a fraction of humans.

So Where Does This Leave Us?

The honest answer: in a much better place than a decade ago, but still miles from the finish line. We can now read nearly the entire genome, detect structural variants and repeat expansions that were invisible before, and layer on transcriptomic and proteomic data to figure out what variants actually do. The review advocates a stepwise approach - start with whole-genome sequencing, follow up with long-read sequencing for complex cases, and keep reanalyzing data as knowledge evolves.

The promise is real: more accurate diagnoses, biologically informed treatments, and maybe - just maybe - a future where we catch neurodegeneration before it catches us.

References

1. Grassano M, Schindler AB, Traynor BJ, Scholz SW. Genetic analysis of neurodegenerative diseases. J Clin Invest. 2026;136(7). DOI: 10.1172/JCI199840

2. Balendra R, Isaacs AM. C9orf72-mediated ALS and FTD: multiple pathways to disease. Nat Neurosci. 2018;21(10):1345-1353. DOI: 10.1038/s41593-018-0241-x. PMCID: PMC6417666

3. Liu Z, Song SY. Genomic and transcriptomic approaches advance the diagnosis and prognosis of neurodegenerative diseases. Genes (Basel). 2025;16(2):135. DOI: 10.3390/genes16020135. PMCID: PMC11855287

4. Firdaus Z, Li X. Unraveling the genetic landscape of neurological disorders: insights into pathogenesis, techniques for variant identification, and therapeutic approaches. Int J Mol Sci. 2024;25(4):2320. DOI: 10.3390/ijms25042320. PMCID: PMC10889342

5. Crary JF. Neurodegeneration: 2024 update. Free Neuropathol. 2024;5:31. DOI: 10.17879/freeneuropathology-2024-5848. PMCID: PMC11736941

Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.