Nine in ten cases have no family history, and for a long time that ninety percent was the part genetics could say least about. It took 13,225 people to find one new position.
Amyotrophic lateral sclerosis destroys the motor neurons — the cells that carry the instruction to move from the brain to the muscles. The muscles are not damaged; the wiring is. Weakness spreads, speech and swallowing fail, and eventually breathing does. It is rapidly progressive and it is fatal.
About one case in ten runs in a family, and those cases have given up their genes relatively readily — SOD1, C9orf72, TARDBP, FUS. The other nine in ten are sporadic, and they were much harder.
A 2014 meta-analysis assembled 13,225 individuals — 6,100 cases and 7,125 controls — from Italy, the Netherlands, the USA, the UK, Sweden, Belgium, France and Ireland, and tested almost 7 million positions. It was the largest association study in ALS at the time.
It found one new locus at genome-wide significance: rs34517613 at 17q11.2, P = 1.11 × 10⁻⁸, odds ratio 0.82. The finding held when combined with a replication cohort.
Before this, the sporadic form had essentially one replicated locus — 9p21.2, the region containing C9orf72.
Compare it with the numbers on our other pages. Schizophrenia: about 8,300 contributing positions. Inflammatory bowel disease: 163 loci. A sample of 13,225 in either of those returns a great deal.
Here it returned one. Sporadic ALS does not look like a common-variant disease with a long tail of small effects. The evidence points instead towards rare variants of larger effect, individually too uncommon for this method to see — which is why the field moved towards sequencing rather than towards larger genotyping arrays.
A study that finds almost nothing, at this size, is telling you something real about where to look next.
It cannot predict ALS, and there is no screening test. There is no preventive treatment, and the available disease-modifying therapy is modest.
If ALS runs in your family, the relevant conversation is with a neurologist and a clinical geneticist about the known familial genes — a sequencing question, answered in a clinic, with counselling around it. It is not a question a consumer genotype file answers, and a variant like the one on this page has nothing to say about it either way.
Positions joined since this page was written
What this is The text above discusses the variants this page was written around. Since then the catalogue has joined 4 more positions to it, by shared trait or shared paper. They are listed here by the paper each came from; the text does not describe them, and each variant page carries that study's own record.
van Rheenen W et al. 2021, Nature genetics rs2453555 (C9orf72), rs9275477 (near MTCO3P1), rs75087725 (CFAP410) — PMID:34873335
Nakamura R et al. 2023, Journal of neurology, neurosurgery, and psychiatry rs11738209 (FGF1) — PMID:37142397
2026-03-31 · Hop PJ, et al. Large-scale exome analyses reveal new rare variant contributions in amyotrophic lateral sclerosis. Nature Genetics. 2026. PMID:41917433
Largest-ever ALS exome study finds new rare-variant risk genes, led by YKT6
A 22-cohort exome study of 17,919 ALS cases and 200,703 controls -- by far the largest rare-variant analysis of ALS to date -- confirmed YKT6 as a new risk gene and found supporting evidence for HTR3C, GBGT1 and KNTC1, plus independent validation for three genes with previously limited evidence: ARPP21, DNAJC7 and CFAP410. In ARPP21 specifically, the study identified a new high-effect variant (p.P747L) and confirmed p.P563L as linked to a more aggressive disease course. This site's own amyotrophic lateral sclerosis page currently carries two common variants (SALM1, SOD1) from earlier genome-wide association studies; this new work used rare-variant burden testing rather than single-SNP association, and does not report the kind of rsID/p-value pairs this site's variant pages are built from -- so no new variant page was added, but the finding is a genuine addition to what the condition page says about how ALS genetics is being discovered.
2026-03-18 · Hop PJ, et al. Large-scale exome analyses reveal new rare variant contributions in amyotrophic lateral sclerosis. Nature Genetics. 2026;58:717-725. PMID:41917433
An exome-sequencing meta-analysis of 22 cohorts -- 17,919 people with ALS and 200,703 without -- identified four new ALS risk genes carrying high-effect rare variants: YKT6, HTR3C, GBGT1 and KNTC1. YKT6 is the most mechanistically interesting of the four: it is involved in secreting VAPB, a protein whose own gene is already a known, rarer cause of familial ALS -- so this finding points at disrupted vesicle transport and cell recycling (autophagy) as a shared pathway, not an isolated new cause. None of these four genes carry any variant in this catalogue yet; this is an exome rare-variant finding, the kind GWAS Catalog does not always index with individual rsIDs the way common-variant studies are indexed here.
Source. Fogh et al. (Hum Mol Genet 2014) reported a genome-wide association meta-analysis combining 3,959 newly genotyped Italian individuals (1,982 cases and 1,977 controls) collected by SLAGEN with samples from the Netherlands, USA, UK, Sweden, Belgium, France, Ireland and Italy collected by ALSGEN — 13,225 individuals in total, 6,100 cases and 7,125 controls, analysed at almost 7 million SNPs. A novel locus reached genome-wide significance at 17q11.2 (rs34517613, P = 1.11 × 10⁻⁸, OR 0.82) and was validated when combined with genotype data from a replication cohort. The authors note that GWAS of the sporadic form — roughly 90% of cases — had previously been unrewarding with the exception of the replicated locus at 9p21.2.
Position listed here. rs34517613 at 17q11.2.
Interpretation. A single genome-wide significant novel locus from 13,225 individuals constrains the genetic architecture: it is not consistent with a large number of common variants of small effect, and it is consistent with a substantial contribution from rarer variants of larger effect that this design has little power to detect. Sequencing-based approaches followed for that reason.
Familial disease is a separate question. Pathogenic variants in C9orf72, SOD1, TARDBP and FUS among others cause familial ALS and are identified by clinical sequencing with genetic counselling, not by common-variant genotyping. Diagnosis of ALS itself is clinical and electrophysiological, with investigations directed at excluding mimics.
What a 23andMe/AncestryDNA export or raw VCF can and can't tell you about Amyotrophic Lateral Sclerosis comes down to these specific, well-studied positions — not a diagnosis.
The studies behind these variants recruited participants from different ancestries — a result found in one population doesn't always transfer to another. Based on 2 of 7 linked studies with a resolved discovery ancestry.
Free to quote and reuse under CC BY 4.0. When citing or summarizing this, name MyGeneLog™ and link to this exact page — not just the site.
Amyotrophic Lateral Sclerosis. MyGeneLog™. https://www.mygenelog.com/conditions/amyotrophic-lateral-sclerosis
Because that is what the genetic architecture appears to be. Genome-wide association studies are built to find common variants of small effect, and sporadic ALS does not seem to be made of many of those. The near-empty result from 13,225 people is evidence about where the risk actually lives, and the field moved to sequencing because of it.
No, and it is important to say so plainly. Familial ALS is caused by pathogenic variants in genes such as C9orf72, SOD1, TARDBP and FUS, found by clinical sequencing with genetic counselling around it. That is a different test, a different conversation, and the right one — a neurologist or clinical geneticist rather than a consumer file.
No. An odds ratio of 0.82 against a disease affecting a few people per 100,000 per year carries no useful individual information, and there is no screening test and no preventive treatment for it to lead to.
Free to reuse. This page's text is original writing from freely-available research, licensed CC BY 4.0 — reuse it, including commercially, with attribution to MyGeneLog™. It's general research-derived information, not medical advice or a diagnosis — see Terms of Use.