Applying Machine Learning for Early Detection of Autism Spectrum Disorder from Maternal Microbiome Data
Bibliographic record
Abstract
Recent research has shown that those who have an inherited predisposition to autism spectrum disorder (ASD) demonstrate a strong association between gut microbiome dysbiosis and brain development through the gut-brain axis. Previous studies demonstrate that this predisposition is predominantly vertically transmitted from mother to child during labour and through breast milk. During this transmission step, several bacterial strains acting as biomarkers in the maternal gut microbiome have been shown to be conserved in the child. Current ASD diagnostic methods rely on behavioral assessments and genetic screening, often delaying intervention and placing additional stress on mothers during early parenthood. Utilizing maternal gut microbiome biomarkers in adjunct with genetic screening could provide a non-invasive screening tool that could enable timely risk detection, facilitating early interventions that improve developmental outcomes and reduce parental anxiety. To achieve this, we leverage random forests, a supervised machine learning algorithm that has been shown to accurately classify microbiome samples as originating from children with ASD or neurotypical children. Additionally, the bacterial species Bacteroides, Lachnospira, Anaerobutyricum, and Ruminococcus torques have been identified as strong biomarkers for ASD. The aim of the study is to evaluate the predictive accuracy of maternal gut microbiome data for pre screening for children with ASD, and validate whether previously identified biomarkers for ASD are conserved in maternal gut microbiome samples. Using samples from a mother’s gut microbiome to determine risk could revolutionize early ASD identification and diagnosis. Considering the connection between genetics and microbiome composition, this approach strengthens traditional genetic screening, offering a more comprehensive risk assessment. Further, by preparing parents earlier in their pregnancies, they can sooner explore supportive therapies, adjust prenatal and postnatal care, and access support systems. This proactive strategy could ultimately improve developmental outcomes, reduce diagnostic delays, and contribute to a deeper understanding of ASD’s environmental and biological influences.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.010 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".