Premium
Advancing biomarker research: utilizing ‘Big Data’ approaches for the characterization and prevention of bipolar disorder
Author(s) -
McIntyre Roger S,
Cha Danielle S,
Jerrell Jeanette M,
Swardfager Walter,
Kim Rachael D,
Costa Leonardo G,
Baskaran Anusha,
Soczynska Joanna K,
Woldeyohannes Hanna O,
Mansur Rodrigo B,
Brietzke Elisa,
Powell Alissa M,
Gallaugher Ashley,
Kudlow Paul,
KaidanovichBeilin Oksana,
Alsuwaidan Mohammad
Publication year - 2014
Publication title -
bipolar disorders
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.285
H-Index - 129
eISSN - 1399-5618
pISSN - 1398-5647
DOI - 10.1111/bdi.12162
Subject(s) - neurocognitive , prodrome , bipolar disorder , biomarker , personalized medicine , omics , genomics , big data , medicine , computer science , psychology , psychiatry , bioinformatics , cognition , biology , genetics , psychosis , genome , gene , operating system
Objective To provide a strategic framework for the prevention of bipolar disorder ( BD ) that incorporates a ‘Big Data’ approach to risk assessment for BD . Methods Computerized databases (e.g., Pubmed, PsychInfo, and MedlinePlus) were used to access English‐language articles published between 1966 and 2012 with the search terms bipolar disorder , prodrome , ‘Big Data’ , and biomarkers cross‐referenced with genomics/genetics , transcriptomics , proteomics , metabolomics , inflammation , oxidative stress , neurotrophic factors , cytokines , cognition , neurocognition , and neuroimaging . Papers were selected from the initial search if the primary outcome(s) of interest was (were) categorized in any of the following domains: (i) ‘omics’ (e.g., genomics), (ii) molecular, (iii) neuroimaging, and (iv) neurocognitive. Results The current strategic approach to identifying individuals at risk for BD , with an emphasis on phenotypic information and family history, has insufficient predictive validity and is clinically inadequate. The heterogeneous clinical presentation of BD , as well as its pathoetiological complexity, suggests that it is unlikely that a single biomarker (or an exclusive biomarker approach) will sufficiently augment currently inadequate phenotypic‐centric prediction models. We propose a ‘Big Data’‐ bioinformatics approach that integrates vast and complex phenotypic, anamnestic, behavioral, family, and personal ‘omics’ profiling. Bioinformatic processing approaches, utilizing cloud‐ and grid‐enabled computing, are now capable of analyzing data on the order of tera‐, peta‐, and exabytes, providing hitherto unheard of opportunities to fundamentally revolutionize how psychiatric disorders are predicted, prevented, and treated. High‐throughput networks dedicated to research on, and the treatment of, BD, integrating both adult and younger populations, will be essential to sufficiently enroll adequate samples of individuals across the neurodevelopmental trajectory in studies to enable the characterization and prevention of this heterogeneous disorder. Conclusions Advances in bioinformatics using a ‘Big Data’ approach provide an opportunity for novel insights regarding the pathoetiology of BD . The coordinated integration of research centers, inclusive of mixed‐age populations, is a promising strategic direction for advancing this line of neuropsychiatric research.