Hoshik Kwon, Su-In Yoon, Sooyeon Baek, Yerin Hyun, Ju Yeon Kim, Dong Hui Park, Jiyoung Yu, Jin Ah Cho, Hyo-Won Kim, Kyunggon Kim
{"title":"Integrated analysis of nutrition profiles and dried blood spot proteomics reveals patterns associated with autism spectrum disorder.","authors":"Hoshik Kwon, Su-In Yoon, Sooyeon Baek, Yerin Hyun, Ju Yeon Kim, Dong Hui Park, Jiyoung Yu, Jin Ah Cho, Hyo-Won Kim, Kyunggon Kim","doi":"10.1080/1028415X.2026.2684633","DOIUrl":null,"url":null,"abstract":"<p><strong>Objectives: </strong>To investigate whether dietary behavior-derived nutritional profiles and proteomic analysis of dried blood spots (DBS) could identify patterns associated with autism spectrum disorder (ASD).</p><p><strong>Methods: </strong>A total of 29 children with ASD (mean age, 56.0 ± 8.7 months; 23 boys) and 27 control children (mean age, 60.6 ± 9.5 months; 20 boys) participated in the study. Three-day food records were used to assess nutrient and food group intakes. Blood samples were collected as DBS. To evaluate group separation, partial least squares discriminant analysis was applied to food group data, nutrient intake, and proteomic data. Balanced accuracy (average of sensitivity and specificity) was estimated using 5-fold cross-validation (100 repeats). Candidate markers were selected based on variable importance in projection (VIP) scores, followed by pathway enrichment analysis of proteomic candidates. Tripartite relationships among candidate nutritional and proteomic markers and clinical measures (Childhood Autism Rating Scale; Aberrant Behavior Checklist; Social Responsiveness Scale; Full-Scale IQ; Vineland Adaptive Behavior Scales) were explored using correlation heatmaps.</p><p><strong>Results: </strong>We prioritized 18 candidate nutritional markers and 10 proteomic markers. The proteins were significantly enriched in nervous system pathways. Correlation analyses revealed several tripartite relationships among nutritional markers, proteomic markers, and clinical measures, including an S100A6-(omega-6 intake or meat intake)-clinical measures relationship and an ALDH18A1-nuts/seeds intake-clinical measures relationship.</p><p><strong>Discussion: </strong>Specific combinations of nutritional profiles and DBS-derived proteomic signatures were associated with symptom severity and functional outcomes in children with ASD. Dietary intake may help interpret these molecular and clinical patterns. Future studies are necessary to elucidate these interrelationships.</p>","PeriodicalId":19423,"journal":{"name":"Nutritional Neuroscience","volume":" ","pages":"1-16"},"PeriodicalIF":3.1000,"publicationDate":"2026-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Nutritional Neuroscience","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1080/1028415X.2026.2684633","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"NEUROSCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
Objectives: To investigate whether dietary behavior-derived nutritional profiles and proteomic analysis of dried blood spots (DBS) could identify patterns associated with autism spectrum disorder (ASD).
Methods: A total of 29 children with ASD (mean age, 56.0 ± 8.7 months; 23 boys) and 27 control children (mean age, 60.6 ± 9.5 months; 20 boys) participated in the study. Three-day food records were used to assess nutrient and food group intakes. Blood samples were collected as DBS. To evaluate group separation, partial least squares discriminant analysis was applied to food group data, nutrient intake, and proteomic data. Balanced accuracy (average of sensitivity and specificity) was estimated using 5-fold cross-validation (100 repeats). Candidate markers were selected based on variable importance in projection (VIP) scores, followed by pathway enrichment analysis of proteomic candidates. Tripartite relationships among candidate nutritional and proteomic markers and clinical measures (Childhood Autism Rating Scale; Aberrant Behavior Checklist; Social Responsiveness Scale; Full-Scale IQ; Vineland Adaptive Behavior Scales) were explored using correlation heatmaps.
Results: We prioritized 18 candidate nutritional markers and 10 proteomic markers. The proteins were significantly enriched in nervous system pathways. Correlation analyses revealed several tripartite relationships among nutritional markers, proteomic markers, and clinical measures, including an S100A6-(omega-6 intake or meat intake)-clinical measures relationship and an ALDH18A1-nuts/seeds intake-clinical measures relationship.
Discussion: Specific combinations of nutritional profiles and DBS-derived proteomic signatures were associated with symptom severity and functional outcomes in children with ASD. Dietary intake may help interpret these molecular and clinical patterns. Future studies are necessary to elucidate these interrelationships.
期刊介绍：
Nutritional Neuroscience is an international, interdisciplinary broad-based, online journal for reporting both basic and clinical research in the field of nutrition that relates to the central and peripheral nervous system. Studies may include the role of different components of normal diet (protein, carbohydrate, fat, moderate use of alcohol, etc.), dietary supplements (minerals, vitamins, hormones, herbs, etc.), and food additives (artificial flavours, colours, sweeteners, etc.) on neurochemistry, neurobiology, and behavioural biology of all vertebrate and invertebrate organisms. Ideally this journal will serve as a forum for neuroscientists, nutritionists, neurologists, psychiatrists, and those interested in preventive medicine.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
