Citation
Pauly, R., Johnson, L., Feltus, F.A., & Casanova, E.L. (2024). “Enrichment of a subset of Neanderthal polymorphisms in autistic probands and siblings.” Molecular Psychiatry, 29, 3452–3461. doi:10.1038/s41380-024-02593-7. Open Access.
Key findings
- Rare Neanderthal-derived variants are significantly enriched in autistic probands compared to race-matched controls, using data from the SPARK cohort (the same large cohort as Litman, Sauerwald et al. 2025 — Genetic programs underlying autism phenotypic heterogeneity), the Genotype-Tissue Expression (GTEx) database, and the 1000 Genomes Project.
- 25 specific SNPs (rare and common) were identified as significantly enriched in autism across different ethnic backgrounds (Black non-Hispanic, White Hispanic, White non-Hispanic). Some show significant clinical associations.
- Genotype-phenotype correlations were identified for SNPs associated with comorbid features including intellectual disability, epilepsy, and language regression — i.e. these Neanderthal-derived variants don’t just contribute to autism diagnosis, they influence the kind of autism someone has.
- Enrichment of Neanderthal DNA is associated with altered brain connectivity patterns (previous work cited): enhanced connectivity within visual processing systems (intraparietal sulcus to occipital cortex and fusiform gyrus) and decreased connectivity within the default mode (social) network. This connectivity pattern is strikingly reminiscent of autistic neural architecture.
- The enrichment holds in sibling analysis — autistic siblings share Neanderthal-derived risk variants at higher rates than expected.
Method in brief
Genome-wide analysis comparing Neanderthal-derived SNP frequencies between autistic probands and race-matched controls. Three cohorts analysed separately (Black non-Hispanic, White Hispanic, White non-Hispanic) to control for population stratification. Both rare-variant enrichment testing and individual SNP-level association were performed. Clinical phenotype correlations used available SPARK phenotype data.
Relevance to this wiki
This paper’s relevance is conceptual rather than immediately practical — but the conceptual work it does is significant:
-
It grounds autistic traits in evolutionary biology. If Neanderthal-derived variants contribute to autism susceptibility, then the neural architecture underlying autism is not a modern anomaly but part of the deep evolutionary history of human cognition. This is fully consistent with a neurodiversity-first approach.
-
The connectivity patterns matter for sensory processing. Enhanced visual processing connectivity and reduced default-mode-network connectivity is a neural signature that maps onto both the heightened sensory detail processing and the social-communication differences that define autism clinically. The paper links Neanderthal DNA to exactly the kind of brain architecture that produces atypical sensory processing.
-
The comorbid phenotype associations are directly relevant. Intellectual disability, epilepsy, and language regression are features of autism with intellectual disability. Finding that some of these associations trace to Neanderthal-derived variants adds a layer to the genetic picture that Litman, Sauerwald et al. 2025 — Genetic programs underlying autism phenotypic heterogeneity paints.
-
It reframes “deficit” as “inheritance.” A trait that has been carried from archaic humans through 50,000 years of introgression and natural selection is not easily described as a defect. It may once have been — and may still be in some contexts — adaptive. This framing aligns with the rejection of deficit language used here (see Positive aspects of hypo- and hyperstimulation).
Limitations
- SPARK cohort is US-based; population stratification is handled but not eliminated.
- “Neanderthal-derived” does not mean “Neanderthal-caused.” Introgressed variants interact with the rest of the modern human genome in complex ways. The enrichment is statistical, not mechanistic.
- The paper identifies associations, not causal pathways. Which of the 25 SNPs actually contribute to autism biology versus being in linkage disequilibrium with something that does is unresolved.
- The paper does not address sensory processing directly — the connectivity findings are cited from prior work, not replicated here.