My long-term research goal is to identify biomarkers of clinical heterogeneity in diseases. The genetic basis of many complex diseases has remained elusive due to inconsistent findings despite the advances of genomic analytic technologies. The failure to replicate the original findings can be attributable to genetic/clinical heterogeneity, gene-gene and gene-environment interactions, to name a few of them. I propose to tackle these challenges with at least two primary approaches: (1) Developing state-of-the-art computational and bioinformatics methods to handle high-dimension data (e.g., combinatorial effects of multiple genetic pathways, environmental factors, and multi-level mediating factors), and (2) generating experiment paradigms to evaluate context-dependent states or traits – of which the variation may be enhanced by the environmental stimuli. My collaborators and I have launched different projects that implement multi-modal research approaches, such as DNA/RNA deep sequencing, functional MRI, magnetic resonance spectroscopy, eye tracking tasks, and various behavioral assays. These measurements may serve as either endophenotypes or modifiers of genetic effects. We anticipate that these approaches can generate findings to shed some mechanistic insight into the emotional and behavioral disturbances. One of my current research foci is autism spectrum disorder (ASD), a disease characterized by impaired social functions. My ongoing and future research plans, with ASD as a disease model, can be summarized as follows:
1. Gene-environment interactions
1.1. Prenatal exposure to drugs
I have proposed a convergent strategy, which uses bioinformatics searching strategies (e.g., Connectivity Map) to identify genes responsive to several pharmacological compounds (e.g., beta 2 adrenergic receptor agonists), and then evaluate if their variants also interact with the history of prenatal exposure to those compounds in case-only pilot samples. Our preliminary data suggest that several synaptic transmission genes may respond to some adrenergic receptor genes, and hence those pairs of genes and compounds may contribute to the risk of ASD. The results will then be validated in a case-control study. I have also proposed to use interval-censored survival analysis to investigate the impact of these compounds on the neurodevelopmental outcomes, such as speech delay.
1.2. Normalizing glutamate-glutamine/GABA cycling to treat social impairment
Glutamate and GABA statuses in ASD have been elusive due to inconsistent findings. We propose to examine the effect of acamprosate (a glutamate receptor antagonist) on mRNA levels of the glutamate-GABA gene pathway and glutamate, glutamine (a substrate of glutamate), and GABA levels in the medial prefrontal cortex (a brain region involved in social cognition) in BTBR T+ Itpr3tf/J mice (BTBR mice), a genetic mouse model of ASD. BTBR mice exhibit reduced social interactions, low exploratory behavior, unusual vocalizations, and high anxiety. We will focus on ultrasonic vocalization (USV) as a primary behavioral outcome, which may be a proxy for social communications in humans. The overarching goal of the project is to evaluate whether a compound with multiple pharmaceutical effects on glutamate and GABA transmission can rescue social communication deficits.
1.3. The impact of childhood traumatic experience
My preliminary evidence has shown that adolescents with the experience of bullying-victimization might have greater levels of reactive aggression and irritability. I have hence hypothesized that negative peer experience (e.g., bullying-victimization) will cause a heightened response in adolescents with ASD compared to their typically developing peers. To test this hypothesis, I will take genetic predisposing factors into consideration to examine gene-environment interactions that jointly impact emotional regulation. The results will shed some light on the mechanism underlying the link between bullying-victimization and emotion dysregulation specific to ASD. Additionally, we plan to apply the similar study design to study the effect of interactions between childhood abuse and genetic factors (i.e., telomere length and epigenetic changes) on neuropsychiatric outcomes in ASD individuals with traumatic experience compared to others without traumatic experience.
2. Genetic basis of endophenotypes and comorbidities related to ASD
2.1. Age of language acquisition in ASD
Age of language acquisition, is often measured with age of first words or phrases. It is regarded as one of the strongest predictors for prognosis for patients with ASD. I have proposed to target some genes on chromosome 7q in light of previous evidence that suggests that that region may harbor quantitative trait loci linked to age of first words/phrases (Alarcon et al., 2002). My preliminary findings show that the FOXP2 gene polymorphisms might interact with advanced paternal age to influence the age of first phrases (Lin et al., 2012). Shortly following this study, we performed a whole-genome scan and found some extended homozygous haplotypes might be associated with the age of language acquisition – the most significantly findings were obtained at the regions that harbor genes associated with cerebellar functions. Interestingly, the associated regions might harbor the variant that underwent recent positive selection according to our findings (Lin et al., 2013). The next step of this research line is to perform deep sequencing for the regions characterized by extended homozygous haplotypes to validate the presence of rare mutations in linkage disequilibrium with nearby common variants. We will be able to trace the ancestral history of the associated coding variants and test the hypothesis that some trait-related mutations have undergone positive selection.
2.2. Glutamate system and frustrative non-reward in irritability in adolescents with ASD
The goal of this study is to examine the neurobiological aspects of “reactive” irritable behaviors in adolescents with ASD. We hypothesize that the heightened emotional response to frustrative non-reward will increase the risk of irritable mood and behaviors in adolescents with mood dysregulation. More specifically, we construct a theoretical model based on prior evidence and our preliminary data – we hypothesize that the relationship between social impairment levels and irritable behaviors/mood is mediated by increased glutamate concentrations, in the prefrontal-limbic system. Furthermore, we hypothesize that glutamate transporter/receptor genetic variants can modify the effect of social impairment on glutamate concentrations under frustrative non-reward. We will also propose to use EEG gamma-band oscillatory waves to measure region-specific glutamate status as an alternative cost-effective tool. Finally, we propose to use eye-tracking tests to measure visual attention inferred from eye gaze movement patterns to predict reactive irritability. Our preliminary data have suggested that visual attention towards cues of hostility is inversely correlated with aggressive tendencies. Combining all of these multiple components, we will be able to measure the relationships among irritability, glutamate status, prefrontal activities, and glutamate transporter/receptor genes.
2.3. Novel outcome measurements in ASD
Individuals with autism spectrum disorder (ASD) have higher levels of emotional arousal than non-ASD individuals. There has been a lack of reliable and objective measurements for emotional arousal in these patients due to their difficulty in expressing their emotions. Inaccurate assessments of emotional arousal in patients with ASD will hinder effective treatments for associated mood and behavioral disturbances. Our preliminary data show that aggressive tendencies might be inversely associated with eye gaze fixation time on hostile objects in non-ASD individuals. Additionally, gamma-band wave activities recorded with encephalogram (EEG) may be correlated with both visual attention and emotional arousal. Such EEG findings may also correlate with glutamatergic transmission in the brain. Therefore, we propose to evaluate the role of visual attention in emotional arousal. We hypothesize that visual attention towards hostile objects is a biomarker for emotional arousal in ASD individuals. We further hypothesize that glutamatergic status (reflected by gamma-band activities) can mediate the link between visual attention and emotional arousal.
In summary, we need epidemiological studies to provide some macro-level insight, which further guides the micro-level biological investigations with other research approaches. On the other hand, we need micro-level biological evidence to provide clues to modifiable risk factors in epidemiologic studies, and provide a better understanding of the disease impact on the society. I will continue to serve as a quantitative data scientist to facilitate biomedical projects that will pave the way for personalized or precision medicine.
Clinical Instructor: The continuing education program Epidemiology of psychiatric
Disorders, I-Ling Hospital, Tainan, Taiwan, 2003-2004.
Teaching Assistant: The course of Introduction to Behavioral and Psychiatric Genetics,
Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland, U.S.A.
Instructor: Peter Zandi, Ph.D., 2003 – 2004 3rd quarter.
Instructor: The Training Course in Statistical Genetic Analysis, Center for Human Genetics, Duke
University Medical Center, Durham, North Carolina, U.S.A., 2005 – 2006
Instructor: The course of Genetic Epidemiology, University of Maryland School of
Medicine, Baltimore, Maryland, USA, 2008
Chief Resident: Providing didactic and bedside teaching activities for medical students and junior residents, National Taiwan University, Taipei, Taiwan, 2011-2012
- PhD student – Cole Brokamp (University of Cincinnati), Gene expression profiles associated with social impairment in autism, CCHMC, 2013
- Master student – Hung-I Hsiao (University of Cincinnati), Gene expression profiles associated with speech delay in autism, CCHMC, 2013
3. Summer undergraduate program - Larry Seifert (Miami University). Interaction of immune-related genes and paternal age in autism spectrum disorder, CCHMC, 2014
4. STAR program for undergraduate students – Sabria Berry (University of Cincinnati), Shared Genetic Profiles of Autism and Medications to Identify Gene-drug Interactions, CCHMC, 2016.
My current collaborative projects are summarized as follows:
1. Shared genetic networks between autism and atopic diseases - collaborator: Tesfeye Marsha, Cincinnati Children's Hospital Medical Center, USA
2. NSAID-triggered genetic functional changes in iPSC-induced neurons associated with autism spectrum disorder - collaborators: Joelle Ruëgg and Carlos Villaescusa, Karolinska Institutet, Sweden
3. Multigenerational epigenetic inheritance: are adverse health effects induced by prenatal chemical exposure transferred to coming generations in humans? - collaborators: Carl-Gustaf Bornehag, Karlstad University, Sweden; Joelle Ruëgg, Karolinska Institutet, Sweden
4. Visual attention in emotional arousal - collaborators: Maria Unenge Hallerback, Karlstad University
My primary research interest is in the genetic basis of clinical heterogeneity in complex disorders. My clinical training includes board certification in adult psychiatry (Taiwan) and visiting scholarship at the Autism Clinic at UCSF (U.S.A.). My research efforts have been devoted to the identification of disease-associated genetic variants using genome-wide association scans and next-generation sequencing. As of now, my publications have been totally cited by 1,368 studies. I have served as a principal investigator on three research projects on the genetics of autism and co-investigator on two NIH R01 projects (genetics of metabolic syndrome and schizophrenia, respectively). My quantitative analytic skills and biomedical knowledge have jointly played a key role in my collaborations with investigators from diverse fields, such as molecular genomics and neuroscience, etc.