|Year : 2021 | Volume
| Issue : 2 | Page : 148-153
Data Analysis of Autism Disorder and Micronutrition using PubMed Abstracts
Seo-Yeong Choe, Dong-Wook Lee, Hwee-Soo Jeong
Department of Family Medicine, Dongguk University College of Medicine, Gyeongju, South Korea
|Date of Submission||01-Nov-2020|
|Date of Decision||22-Nov-2020|
|Date of Acceptance||16-Dec-2020|
|Date of Web Publication||22-Apr-2021|
Department of Family Medicine, Dongguk University College of Medicine, Dongda-ro 123, Gyeongjusi, Gyeongsangbuk-do 38066
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Aim: This study was conducted to identify the notable nutrients mentioned in abstracts of articles about autism disorder using the big data analysis method. Methods and Materials: Abstracts of articles on nutrients related to autism disorder were extracted from MEDLINE using PubMed. Text mining was performed to extract nutrient-related words from collected research abstracts. Network analysis was performed using nutrients keywords and showed more than 100 occurrences among extracted words. Results: As a result of text mining 932 abstracts, the top 10 Nutrients that appeared more than 100 times were vitamin D omega 3, zinc, iron, copper, vitamin B6, vitamin B12, folate, calcium, and probiotics, in decreasing order. Folate and copper were central nutrients in the central analysis of the network made by 10 nutrients. Grouping the 10 nutrients showed folic acid, vitamin B6, vitamin B12, calcium, iron, zinc, copper, omega 3 in group 1, probiotics in group 2, and vitamin D in group 3. Conclusions: A variety of nutrients, including vitamin D, folate, and copper, were found to be related to autism disorder by big data analysis of abstracts of nutrition research studies related to this disorder.
Keywords: autistic disorder, copper, folate, nutrition, vitamin D
|How to cite this article:|
Choe SY, Lee DW, Jeong HS. Data Analysis of Autism Disorder and Micronutrition using PubMed Abstracts. Int J Nutr Pharmacol Neurol Dis 2021;11:148-53
|How to cite this URL:|
Choe SY, Lee DW, Jeong HS. Data Analysis of Autism Disorder and Micronutrition using PubMed Abstracts. Int J Nutr Pharmacol Neurol Dis [serial online] 2021 [cited 2022 Aug 8];11:148-53. Available from: https://www.ijnpnd.com/text.asp?2021/11/2/148/314377
Key Messages: Vitamin D, omega 3, zinc, iron, copper, vitamin B6, vitamin B12, folate, calcium, and probiotics were found to be nutrients related to autism disorder in research articles.
| Introduction|| |
Autism exhibits impairments in communication and social interactions, and those affected have difficulty controlling behavior and emotions. The condition shows a rapid increasing trend worldwide with a median prevalence of 62 per 10,000, and it is diagnosed four to five times more often in boys than girls.,,
Although genetic factors contribute greatly to autism spectrum disorder, environmental factors also contribute, and perinatal factors such as drug use, perinatal infection, and nutritional deficiencies, are also known to directly or indirectly affect the phenotype of autism spectrum disorder., In addition, many children with autism spectrum disorder exhibit abnormal gastrointestinal physiologies such as increased intestinal permeability, overall microbial changes, and intestinal infections. To alleviate gastrointestinal tract and behavioral symptoms, the nutritional intervention has been attempted and supplementation with micronutrients such as folic acid and vitamins has been shown to prevent childbirth with autism and relieve autistic symptoms in children. However, due to the lack of well-designed studies, no consensus has been reached regarding optimal nutritional therapy for autism.
Therefore, this study was conducted to identify nutrients mentioned in abstracts of autism disorder extracted from PubMed using the big data analysis method, and to suggest research directions for nutritional intervention in autism disorder.
| Subjects and Methods|| |
On April 24, 2020, the authors used PubMed to extract abstracts of articles on nutrients related to autism disorder in MEDLINE. Medical Subject Headings (MeSH) terms such as “Autistic Disorder,” “Kanner’s Syndrome,” “Kanner Syndrome,” “Infantile Autism,” and “Autism” were used as search terms, and “Dietary Supplement,” “Food Supplements,” “Nutriceutical,” “Neutraceutical,” “Herbal Supplement,” “Nutrient,” “Macronutrient,” “Micronutrient,” “Trace elements,” “Biometal,” “Trace mineral,” and “Vitamin” were used. The publication period was set from January 1, 1900 to December 31, 2019, and the search was limited to articles written in English. The final search formula used was “(Autistic Disorder OR Kanner’s Syndrome OR Kanner Syndrome OR Infantile Autism OR Autism OR Early Infantile Autism) AND (Dietary Supplement OR Food Supplements OR Nutriceutical OR Neutraceutical OR Herbal Supplement OR nutrient OR macronutrient OR micronutrient OR trace elements OR biometal OR trace mineral OR vitamin) Filters: Abstract; Publication date to 2019/12/31; English”. After downloading the extracted research abstracts as XML files, they were converted into Excel files using PubMed2XL (Ver 2.01). The converted Excel file included PubMed ID (PMID), Digital Object ID, Article Title, Abstract, Language, Journal: Title, Journal: Abbreviation, Journal: Year, Citation Medium, First Author, All Authors, and Affiliation.
Text mining was performed to extract nutrient-related words from the collected research abstracts. Text mining is a method of extracting keywords from a large number of text documents. To extract nutrient keywords from collected abstracts, the authors extracted only nouns in primary text mining, and then two authors selected nutrient words among nouns. After unifying nutrient words by creating a synonym list, secondary text mining was conducted to select the final nutrient keywords. Network analysis was performed using nutrients keywords showing more than 100 occurrences among extracted words [Figure 1].
Network analysis is used in human relations research for the first time and is increasingly being used in the natural sciences and the medicinal field, and by policymakers and social media. By analyzing networks consisting of nodes with individual properties and links between them, central nodes in networks and subgroups can be identified. In this study, a two-mode network in the form of “abstracts × keywords” was converted into a one-mode network as a “keyword × keyword” based on the co-occurrence by cosine similarity. In order to identify keywords central to the network, we calculated “degree,” “closeness,” and “betweenness” centrality scores as well as Eigenvector and PageRank centrality, which were calculated by considering degrees of connection between central nutrients. Community analysis was conducted to identify subgroups in the network by removing weak links. Text mining and network analysis were analyzed using Netminer Version 4.0 (Netminer, Cyram Inc., Seoul).
| Results|| |
The total number of abstracts of research articles extracted using PubMed was 932. Abstracts were first published from 1975 and fewer than 10 articles per year were published until 2005, and then numbers published increased to 10 to 50 articles per year during 2006 to 2012, 50 to 100 articles from 2013 to 2016, and more than 100 articles from 2017 [Figure 2].
As a result of text mining all abstracts, 7612 noun-type keywords were extracted during first text mining, and 71 nutrient keywords were extracted during second mining. Of these keywords, the top 10 nutrients appeared more than 100 times; vitamin D appeared 309 times followed in decreasing order by omega 3, zinc, iron, copper, vitamin B6, vitamin B12, folate, calcium, and probiotics. Considering differences between numbers of abstracts published per annum, vitamin B6 appeared most in abstracts from 1975 to 2005, omega 3 from 2006 to 2012, zinc from 2013 to 2016, and vitamin D from 2017 to 2019 [Table 1].
|Table 1 Top 10 nutrients that have the most frequent occurrences in research articles|
Click here to view
The network of top10 nutrients consisted of 10 nodes (nutrients) and 35 links, and network density was 0.778 (range 0–1). The average score of degree between nutrients was 3.5, that is, one nutrient was linked to three to four nutrients in the network. The average distance between nutrients was 1.22, and nutrients were connected in 1.2 steps. The network is shown in [Figure 3].
|Figure 3 The network of top 10 nutrients associated with autism disorder. The node represents individual nutrients. The higher frequency of appearance in articles, the larger size of the node. The higher frequency of co-appearance between two nodes in articles, the thicker width of the link|
Click here to view
Centrality analysis of five methods identified folate as the central nutrient in the network, which had a high score in most centrality analysis results, except for Eigenvector centrality. [Table 2]. Folate had a connection relationship with nine other nutrients, and distances and paths between nutrients were short. However, Eigenvector centrality analysis showed copper had the highest score of centrality at 0.58 followed by zinc (0.57). As a result of community analysis, the network was divided into three subgroups. Group 1 contained folate, vitamin B6, vitamin B12, calcium, iron, zinc, copper, and omega 3, Group 2 contained probiotics, and Group 3 contained vitamin D [Figure 4].
|Table 2 Centrality scores of nutrients in five centrality analyses of network|
Click here to view
|Figure 4 Subgroups in the network. Group 1 contained folate, vitamin B6, vitamin B12, calcium, iron, zinc, copper, and omega 3, Group 2 contained probiotics, and Group 3 contained vitamin D|
Click here to view
| Discussion|| |
Given the many medical studies being published, unstructured data analysis tools such as text mining and social network analysis are being used to extracting knowledge from the literature., This study was used to identify nutrients related to autism from the abstracts of published papers.
In 932 abstracts on autism disorder and nutrition, the top 10 nutrients in terms of frequent appearances identified by text mining were vitamin D, omega 3, zinc, iron, copper, vitamin B6, vitamin B12, folate, calcium, and probiotics in decreasing order.
Vitamin D was the most frequently mentioned nutrient, and vitamin D status is known with scientific certainty that it is related to autism disorder, as children with autism exhibit vitamin D deficiency. Furthermore, observational studies reported to be vitamin D supplementation improves symptoms in children with autism disorder., Based on these results, a large-scale randomized study was suggested to investigate the effects of vitamin D on prevention and symptom relief. Supplementation of omega 3, an unsaturated fatty acid essential for brain development, mood, and behavioral development, was also reported to improve the symptoms of autism disorder., Perinatal zinc deficiency is a known risk factor of autism disorder, and zinc concentrations in the blood and saliva of children with autism are significantly lower than normal.,, In addition, iron deficiency has been associated with neurobehavioral diseases such as autism disorder and attention deficit hyperactivity disorder (ADHD).
Network analysis showed the most central nutrient was folate. Folate is well known to prevent neural tube defects and to be a major donor nutrient of methylene tetrahydrofolate reductase, which is involved in the synthesis of S-adenosyl methionine. Among the polymorphisms of this enzyme, MTHFR C677T and MTHFR A1298C have been reported to increase the risk of autism. In a population study, folate intake during the perinatal period was found to be associated with a low autistic infant birthrate. Recently, studies have been undertaken on the association between low folate concentration in cerebrospinal fluid (called cerebral folate deficiency), folate receptor autoantibodies, and autism disorder., In addition to omega 3, zinc, iron, and vitamin B6 and B12, calcium was also identified as a central player in the network. Vitamins B6 and B12 have been reported to effectively improve the symptoms of autism,, and calcium supplementation was found to be helpful when administered with vitamin D.
Copper only showed a high score in Eigenvector centrality analysis, that is, in one of the five centrality analyses conducted. Unlike conventional analyses of degree, closeness, and betweenness, which are based on a simple relationship, Eigenvector centrality shows a higher score when a nutrient with high centrality has many connections. Copper is thought to have a high centrality because it was linked only to nutrients with high centrality values, such as zinc, iron, and calcium. On the other hand, folate was considered to have a relatively low centrality value because it was linked to all nutrients. However, folate was confirmed as the most central nutrient by Pagerank centrality analysis, which is similar to Eigenvector analysis. Copper has neurotoxic effects and is considered to be associated with zinc deficiency, and thus, may influence the development of autism disorder symptoms when Zn/Cu ratio in serum is low.,
As a result of dividing the network into subgroups, the subgroup consisting of seven nutrients including folate was the largest. Vitamin D and probiotics were classified as each individual subgroup, which could be explained that articles on vitamin D or probiotics, unlike other nutrients, were reported as single research topics. In recent research papers, probiotics appear to be mentioned more often. However, a systematic review of the effects of probiotics on the symptoms of autism disorder suggested their effects are limited.
This study has several limitations. First, it was conducted using abstracts related to autism disorder and nutrition extracted using PubMed MeSH terms, and thus, there is a possibility that some research abstracts were missed. On the other hand, there is a possibility of the same error if research abstracts not stored in MEDLINE were added.
Second, two researchers participated in the second text mining, extracted nutrient words, and agreed on nutrient words through cross-check, but human errors that may have occurred during this process should be considered. Third, this study does not be shown the mechanisms responsible for the activities of nutrients in autism disorder. Nevertheless, this study is meaningful as the first study to identify central micronutrients using a big data analysis method with autism research abstracts, and further studies on the effects of micronutrients identified in this study on autism disorders will be proposed.
| Conclusion|| |
Vitamin D, omega 3, zinc, iron, copper, vitamin B6, vitamin B12, folate, calcium, and probiotics were found to be notably associated with autism by big data analysis of abstracts on autism disorder and nutrition research.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Lord C, Brugha TS, Charman T et al.
Autism spectrum disorder. Nat Rev Dis Primers 2020;6:5.
Lai MC, Lombardo MV, Baron-Cohen S. Autism. Lancet 2014;383:896-910.
Elsabbagh M, Divan G, Koh YJ et al.
Global prevalence of autism and other pervasive developmental disorders. Autism Res 2012;5:160-79.
Tordjman S, Somogyi E, Coulon N et al.
Gene × Environment interactions in autism spectrum disorders: role of epigenetic mechanisms. Front Psychiatry 2014;5:53.
Principi N, Esposito S. Vitamin D deficiency during pregnancy and autism spectrum disorders development. Front Psychiatry 2020;10:987.
Ristori MV, Quagliariello A, Reddel S et al.
Autism, gastrointestinal symptoms and modulation of gut microbiota by nutritional interventions. Nutrients 2019;11:2812.
Tan M, Yang T, Zhu J et al.
Maternal folic acid and micronutrient supplementation is associated with vitamin levels and symptoms in children with autism spectrum disorders. Reprod Toxicol 2020;91:109-15.
Sathe N, Andrews JC, McPheeters ML, Warren ZE. Nutritional and dietary interventions for autism spectrum disorder: a systematic review. Pediatrics 2017;139:e20170346.
Isaak D, PubMed2XL (version 2.01). J Med Libr Assoc 2016;104:92-94.
Thompson P, Batista-Navarro RT, Kontonatsios G et al.
Text mining the history of medicine. PLoS One 2016;11:e0144717.
Yan E, Ding Y. Applying centrality measures to impact analysis: a coauthorship network analysis. J Am Soc Inf Sci Technol 2009;60:2107-18.
Franceschet M. PageRank: Standing on the shoulders of giants. Commun ACM 2011;54:91-101.
Gonzalez GH, Tahsin T, Goodale BC, Greene AC, Greene CS. Recent advances and emerging applications in text and data mining for biomedical discovery. Brief Bioinform 2016;17:33-42.
Fernell E, Bejerot S, Westerlund J et al.
Autism spectrum disorder and low vitamin D at birth: a sibling control study. Mol Autism 2015;6:3.
Jia F, Wang B, Shan L, Xu Z, Staal WG, Du L. Core symptoms of autism improved after vitamin D supplementation. Case Rep Pediatr 2015;135:e196-8.
Feng J, Shan L, Du L et al.
Clinical improvement following vitamin D3 supplementation in autism spectrum disorder. Nutr Neurosci 2017;20:284-90.
Mazahery H, Camargo CA Jr, Conlon C, Beck KL, Kruger MC, von Hurst PR. Vitamin D and autism spectrum disorder: a literature review. Nutrients 2016;8:236.
Kidd PM. Omega-3 DHA and EPA for cognition, behavior, and mood: clinical findings and structural-functional synergies with cell membrane phospholipids. Rev Altern Med Rev 2007;12:207-27.
Mazahery H, Conlon C, Beck KL et al.
A randomised-controlled trial of vitamin D and omega-3 long chain polyunsaturated fatty acids in the treatment of core symptoms of autism spectrum disorder in children. J Autism Dev Disord 2019;49:1778-94.
Grabrucker S, Jannetti L, Eckert M et al.
Zinc deficiency dysregulates the synaptic ProSAP/Shank scaffold and might contribute to autism spectrum disorders. Brain 2014;137:137-52.
Crăciun EC, Bjørklund G, Tinkov AA et al.
Evaluation of whole blood zinc and copper levels in children with autism spectrum disorder. Metab Brain Dis 2016;31:887-90.
Deshpande RR, Dungarwal PP, Bagde KK, Thakur PS, Gajjar PM, Kamath AP. Comparative evaluation of salivary zinc concentration in autistic and healthy children in mixed dentition age group-pilot study. Indian J Dent Res 2019;30:43-46.
] [Full text]
Pivina L, Semenova Y, Doşa MD, Dauletyarova M, Bjørklund G. Iron deficiency, cognitive functions, and neurobehavioral disorders in children. Rev J Mol Neurosci 2019;68:1-10.
Mohammad NS, Jain JM, Chintakindi KP, Singh RP, Naik U, Akella RR. Aberrations in folate metabolic pathway and altered susceptibility to autism. Psychiatr Genet 2009;19:171-6.
Levine SZ, Kodesh A, Viktorin A et al.
Association of maternal use of folic acid and multivitamin supplements in the periods before and during pregnancy with the risk of autism spectrum disorder in offspring. JAMA 2018;75:176-84.
Ramaekers V, Sequeira JM, Quadros EV. Clinical recognition and aspects of the cerebral folate deficiency syndromes. Clin Chem Lab Med 2013;51:497-511.
Quadros EV, Sequeira JM, Brown WT et al.
Folate receptor autoantibodies are prevalent in children diagnosed with autism spectrum disorder, their normal siblings and parents. Autism Res 2018;11:707-12.
Sato K. Why is vitamin B6 effective in alleviating the symptoms of autism? Med Hypotheses 2018;115:103-106.
Hendren RL, James SJ, Widjaja F, Lawton B, Rosenblatt A, Bent S. Randomized, placebo-controlled trial of methyl B12 for children with autism. J Child Adolesc Psychopharmacol 2016;26:774-83.
Srinivasan S, O’Rourke J, BerscheGolas S, Neumeyer A, Misra M. Calcium and vitamin D supplement prescribing practices among providers caring for children with autism spectrum disorders: are we addressing bone health? Autism Res Treat 2016;2016:6763205.
Bulcke F, Dringen R, Scheiber IF. Neurotoxicity of copper. Adv Neurobiol 2017;18:313-43.
Ng QX, Loke W, Venkatanarayanan N, Lim DY, Soh AYS, Yeo W. A systematic review of the role of prebiotics and probiotics in autism spectrum disorders. Medicina 2019;55:129.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2]