Obnet: Network of semantic associations for obesity
BMC Bioinformatics volume 15, Article number: P6 (2014)
Obesity is a complex disease that causes health complications and mortality. Recently, the Centers for Disease Control (CDC) classified obesity as a disease and introduced guidelines for diagnosis. To understand complex relationships among all contributing factors, we have developed the ObNet (Network of Semantic Association for Obesity). The concepts were adopted from ARIANA [1, 2]. The Local Latent Semantic Analysis tool , developed in the CVPIA lab, was used to extract network of semantic associations. Web services and visualization were developed to display the associations in a flexible network form that is easy to interact.
Materials and methods
In this pilot study, 198 Medical Subject Headings (MeSH) terms were selected as potential risk factors by a clinician that specializes in weight gain and obesity. Abstracts from the PubMed database were mined to create the semantic model. A Query-Based Sampling (QBS) process  was used to create balanced and “Representative Samples” for modeling the ObNet. A relevance model  was adopted to categorize associations in three groups: Clear Association, Established Concepts, and Current Research Topics.
ObNet is a web service that provides users with the network of semantic associations for obesity, using literature data. The system can be queried with an array of keywords related to obesity to extract a customized network of associated entities. A pilot study using Obnet revealed current research topics in obesity. Three terms were identified and evaluated by a field expert: dietary fat, adults, and attitude to health. Using a direct literature search, corroborating evidence was found to indicate the three identified terms were current research topics. First, Tan et al.  investigated the anti-metabolic disorder effects of kumquat (Fortunella margarita Swingle) fruit extract (FME) on high-fat diet-induced C57BL/6 obese mice. Second, in another study, 149 primary care physicians were surveyed and their knowledge and perceived barriers in the management of overweight and obesity were measured . Finally, in a recent investigation, the impacts of neighborhood park access and quality on body mass index among adults in New York City were measured .
The ObNet is an innovative web service tool for the assessment of obesity. It has the potential to support diagnosis and uncover novel relationships with the disease. Empirical study suggests that the model is capable in finding network of concepts and current research trends for obesity related keywords. Future direction of the study will improve the visualization module, and develop a mobile application.
Abedi V, Zand R, Yeasin M, Faisal FE: An automated framework for hypotheses generation using literature. BioData Min. 2012, 5 (1): 13-10.1186/1756-0381-5-13.
Abedi V, Yeasin M, Zand R: ARIANA: Adaptive robust and integrative analysis for finding novel associations. Proceedings of the 2014 International Conference on Advances in Big Data Analytics: 21-24 July 2014; Las Vegas. 2014,
Muthukuri K: Query based sampling and local semantic analysis to model network association between drugs and diseases. MS thesis. 2013, The University of Memphis, Bioinformatics Program
Tan S, Li M, Ding X, Fan S, Guo L, Gu M, Zhang Y, Feng L, Jiang D, Li Y, Xi W, Huang C, Zhou Z: Effects of fortunella margarita fruit extract on metabolic disorders in high-fat diet-induced obese C57BL/6 mice. PLoS One. 2014, 9 (4): e93510-10.1371/journal.pone.0093510.
Sebiany AM: Primary care physicians’ knowledge and perceived barriers in the management of overweight and obesity. J. Family Community Med. 2013, 20 (3): 147-152. 10.4103/2230-8229.121972.
Stark JH, Neckerman K, Lovasi GS, Quinn J, Weiss CC, Bader MD, Konty K, Harris TG, Rundle A: The impact of neighborhood park access and quality on body mass index among adults in New York City. Prev Med. 2014, 64: 63-68.
This work was supported by the Electrical and Computer Engineering Department and Bioinformatics program at the University of Memphis as well as by NSF grants NSF-IIS-0746790. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the funding institution.
About this article
Cite this article
Taghizad, H., Yeasin, M., Cherry, T. et al. Obnet: Network of semantic associations for obesity. BMC Bioinformatics 15 (Suppl 10), P6 (2014). https://doi.org/10.1186/1471-2105-15-S10-P6