Table of Contents

    Topic review

    Pharmacogenomic Biomarkers in Psychiatry

    View times: 23
    Submitted by: Hotcherl Jeong

    Definition

    Pharmacogenomic biomarkers are potential individual genetic variations that can affect drug response influencing both pharmacokinetic parameters by causing variable activity of the systems responsible for the absorption, distribution, metabolism, and excretion of the drug and pharmacodynamic parameters like the mechanisms of action of the drug. Here, the term "pharmacogenomic biomarkers in psychiatry" means those related to a variety of psychiatric disorders, such as depression, ADHD, narcolepsy, schizophrenia, bipolar disorder, and epilepsy. 

    1. Introduction 

    The World Health Organization (WHO) estimates that about 25% of the population around the world will suffer from at least one mental disorder at some time in their lives [1][2]. Depression and anxiety are among the most common disorders, and these can affect people regardless of age, gender, ethnicity, or background. We do not fully understand what causes most cases of mental health impairment, but it is known that both genetic and environmental factors can often contribute to an individual's predisposition to a particular disorder. In other cases, serious injuries or traumatic events cause psychological symptoms that persist for a long period of time [3].

    Medications can be used in order to reduce the intensity of symptoms or treat several psychiatric disorders. A patient's response to the many medications used to treat various psychiatric disorders can be highly variable [4]. Drug response is dependent on personal health risk factors (e.g., gender, age, liver and renal function, blood pressure, body fat, alcohol and drugs, and drug–drug interactions). In addition, genetic factors, i.e., individual's unique genetic makeup, can affect drug response influencing both pharmacokinetic parameters by causing variable activity of the systems that are responsible for the absorption, distribution, metabolism, and excretion of the drug and pharmacodynamic parameters, like the mechanisms of action of the drug [5][6]. Pharmacogenomics (PGx) refers to the study of drug response as it relates to potential individual genetic variations.

    For an increasing number of drugs, pharmacogenomic testing is available and used to pre-screen patients and help them in selecting drug choice and drug dose accordingly [4][7]. Now, more than 10% of medications that are approved by the U.S. Food and Drug Administration (FDA) provide pharmacogenomic information (PGx information) in their drug labeling. This proportion is gradually increasing as more pharmacogenomic biomarkers (PGx biomarkers) are discovered and validated.

    There are solid reasons for pharmacogenomic testing (PGx testing). Some drugs are only effective for specific genotypes and the testing can avoid unpredictable, severe, and potentially fatal drug reactions. Furthermore, for some drugs, a patient's ancestry is the essential consideration. For example, for carbamazepine, a commonly used antiepileptic drug, the FDA recommends that, if patients are descendants of genetically high-risk populations, they should take PGx testing for the presence of HLA-B*15:02 before treatment [8][9][10]. Carriers of this variant, which is frequently found in Han Chinese descendants, are highly susceptible to the development of Stevens–Johnson syndrome and toxic epidermal necrolysis, which often lead to serious conditions, during the course of carbamazepine therapy. The HLA-B variant alleles are just one example of such adverse drug reactions (ADRs). In fact, there is a plethora of genetic variants that are associated with ADRs. As an evident example, carriers of a variant of MT-RNR1 (mitochondrially encoded 12S rRNA), an RNA-coding gene, are at high risk of irreversible hearing loss by a single dose of gentamicin [11][12].

    For a growing number of drugs, PGx testing provide a means of optimizing the drug choice and drug dose. Drug labels include not only standard dosing information, but also guidelines for adjusting the drug dose or selecting an alternative drug, when necessary, based on a patient's genetic makeup if gene-drug interrelationships are well understood. Dosing adjustment requirements or recommendations are mostly in variants of genes that encode drug-metabolizing enzymes or drug transporters [13]. Thus, PGx biomarkers in genetic variants that are important for interindividual variations in PK and PD have been very useful in the optimization of pharmacotherapy. Several independent institutions, including the FDA [14], the European Medicines Agency (EMA), the Clinical Pharmacogenetics Implementation Consortium (CPIC) [15], the Canadian Pharmacogenomics Network for Drug Safety (CPNDS) [16][17], and the Dutch Pharmacogenetics Working Group (DPWG), have provided instructions on how PGx testing results can be interpreted in terms of the drug choice and the drug dose [18][19][20].

    Accumulated data are then noted to FDA and its Table of Pharmacogenomic Biomarkers in Drug Labeling is widely used as a standard guideline [14]. PGx information is only included on labels when it is useful to inform clinicians of the impact of genotype on phenotype—gene–drug interrelationships—or to indicate whether a PGx test is available for a particular medication. As of now, the Table of PGx Biomarkers includes 431 drug-biomarker pairs for 298 drugs across therapeutic areas. In addition, PharmGKB provides a comprehensive resource, in which evidence-based PGx knowledges are curated and disseminated by scientific team about how our body responds to medications [21]. Pharmacogenomic information is important: it can maximize drug efficacy and reduce/avoid drug toxicity. Currently, FDA's Table of PGx Biomarkers describes PGx information for 35 psychiatric medications, as in Table 1. In addition, the Table of PGx Biomarkers includes PGx information for eight antiepileptic drugs (AEDs), as in Table 2.

    Table 1. Food and Drug Administration (FDA) pharmacogenomic biomarkers in drug labeling in psychiatry.

    H3R antagonist = histamine H3 receptor antagonist; NDRI = norepinephrine-dopamine reuptake inhibitor; SARI = serotonin antagonist and reuptake inhibitor; SNRI = serotonin and norepinephrine reuptake inhibitor; SSRI = selective serotonin reuptake inhibitor; TCA = tricyclic and tetracyclic antidepressant; WPA = wakefulness promoting agent.

    Table 2. FDA pharmacogenomic biomarkers in drug labeling in epilepsy.

    Drug

    Type

    Indication

    Biomarker

    FDA

    FDA Labeling Sections

    EMA

    Brivaracetam

    Inhibits synaptic vesicle SV2A protein

    Epilepsy

    CYP2C19

    Actionable

    Clinical Pharmacology

    Actionable

    Carbamazepine

    Enhances sodium channel (rapid inactivation)

    Inhibits L-type calcium channel

    Epilepsy

    Bipolar disorder

    HLA-B

    Testing Required

    Boxed Warning

    Warnings

    Precautions

     

    HLA-A

    Actionable

    Warnings

     

    Clobazam

    GABAA receptor agonist

    Epilepsy

    CYP2C19

    Actionable

    Dosage and Administration

    Use in Specific Populations

    Clinical Pharmacology

     

    Diazepam

    GABAA receptor agonist

    Epilepsy

    CYP2C19

    Actionable

    Clinical Pharmacology

     

    Lacosamide

    Enhances sodium channel (slow inactivation)

    Epilepsy

    CYP2C19

    Informative

    Clinical Pharmacology

    Informative

    Oxcarbazepine

    Enhances sodium channel (rapid inactivation)

    Inhibits N/P- and R-type calcium channel

    Epilepsy

    Bipolar disorder

    HLA-B

    Testing recommended

    Warnings and Precautions

     

    Phenytoin

     

    Enhances sodium channel (rapid inactivation)

    Epilepsy

    CYP2C9

    Actionable

    Clinical Pharmacology

     

    CYP2C19

    Clinical Pharmacology

    HLA-B

    Warnings

    Valproic Acid

    Inhibits voltage-dependent sodium and T-type calcium channels

    Enhances GABA transmission

    Epilepsy

    POLG

    Testing Required

    Boxed Warning

    Contraindications

    Warnings and Precautions

     

    Nonspecific (Urea cycle disorders)

    Actionable

    Contraindications

    Warnings and Precautions

     

    2. CYP2D6 and CYP2C9 Genes

    The cytochrome P450s (CYPs) comprise a large superfamily of a variety of enzymes that serve as major workhorses for metabolizing steroid hormones, lipids, toxins, and xenobiotics. The CYP superfamily genes encode enzymes that function as monooxygenases and catalyze the modification of about 25–30% commonly used drugs [22][23]. The CYP genes are quite polymorphic and they can lead to increased, decreased, or completely absent drug metabolism activity. Among these genes, CYP2D6 is particularly important and heavily studied. More than 100 CYP2D6 variants have been reported and catalogued at the Pharmacogene Variation Consortium database [24]. In addition to large numbers of single nucleotide polymorphisms (SNPs), other types of variations—gene deletions, duplications, copy-number variants, and pseudogenes that are close to the gene—make genotyping very challenging.

    Many of these variants cause the enzyme to change activity at different levels. The level of CYP2D6 activity decides how an individual responds to the substrate drugs. A standard dosage of the drug may show inadequate efficacy in some individuals and serious toxicity in others. To name a few, the drug substrates of CYP2D6 include atomoxetine (a non-stimulant for ADHD), clozapine (an antipsychotic for schizophrenia), and venlafaxine (an antidepressant), among psychiatric medications, as in Table 1 [4][25]. For these drugs, standard doses will result in higher-than-optimal active levels when individuals have absent or deficient CYP2D6 activity. Thus, the risk of ADRs increases and it may result in treatment failure.

    There are substantial variations in CYP2D6 allele frequencies among different populations [26][27]. The wild-type CYP2D6*1 allele shows normal enzyme activity and the extensive or normal metabolizer phenotype. The CYP2D6*2, -*33, and -*35 alleles also belong to this group. Other alleles contain non-functional variant(s), which produce a non-functioning enzyme (*3, *4, *5, *6, *7, and *8) or a decreased-activity enzyme (*10, *17, *29, and *41) [28]. Intermediate and poor metabolizers are individuals who carry decreased and null CYP2D6 alleles, respectively. Notably, approximately 30% of Asians and Asian descendants are intermediate metabolizers. In these populations, the *10 allele with decreased activity is very common: about 40%, when compared with about 2% in Caucasians [29]. Thus, a large proportion of Asians belong to intermediate metabolizers than Caucasians [30]. The African and African American populations also show a large proportion of CYPD6 alleles having sub-optimal activity. The frequencies of the remaining alleles vary depending on the population [30][31][32]. In Caucasians, only small proportions (less than 10%) are poor metabolizers [30]. In contrast, approximately 40% are extensive/normal metabolizers who carry two copies of *1 allele [33][34][35]. CYP2D6 poor metabolizers show higher levels of amitriptyline (as an example of drug substrates) in the plasma, when compared with extensive metabolizers, after standard doses of amitriptyline are taken [36]. When individuals carry a CYP2D6 null variant, their risk of developing ADRs becomes, at least, moderately increased [37]. Because standard dosages may cause to ADRs in poor metabolizers, it is recommended to avoid many tricyclic antidepressants (TCAs) and, instead, take an alternative option, a drug that is not a substrate of CYP2D6 [38].

    Interestingly, copy-number variants were also found in CYP2D6 genes [24]. In other words, individuals who carry more than two copies of functional CYP2D6 alleles—three to 13 copies of CYP2D6 active allele—have been reported. These carriers are CYP2D6 ultrarapid metabolizers. In the case of 13 functional copies, the rate was up to 17 times higher than for individuals with no active CYP2D6 enzyme [39]. If the drug substrate has increased rate of metabolism, then its active form will not be available and, thus, the therapeutic response will become poor.

    3. Beyond Pharmacogenomic Testing

    Precision medicine targets providing the optimal diagnoses and treatments for each patient based on the categorization of biomarkers [40][41][42][43]. PGx is one of the main research areas of precision medicine. Nowadays, advances in artificial intelligence (AI), machine learning, multi-omics, and neuroimaging allow for analyzing and integrating complex genomic and clinical data in psychiatry and neurology. Artificial intelligence is the field of computing science that produces an algorithm based on available data to create predictive outcomes, even for unknown data in the future [44][45][46][47]. Particularly, state-of-the-art technology of deep learning revolutionized bioinformatics and medical imaging by yielding helpful software tools [48][49]. Whereas cancer therapy has routine clinical settings with well-established genomic data, in neuropsychiatry, the relationship between PGx data and their clinical significances has not been fully studied. Thus, the usage of artificial intelligence remains limited in the field.

    AI has been used in predicting diagnosis, treatment outcome, and prognosis. As of psychiatry and neurology, multiple studies have used models, including deep learning architecture, random forest, tree-based ensemble, elastic net, and linear regression in order to evaluate and predict lithium treatment response on major depressive disorder [50]. To predict prognosis of major depressive disorder, there are algorithms, such as Gaussian process algorithm, Deep Patient, DeepCare, and Doctor AI, which utilize electronic health records. For example, Deep Patient has forecasted psychiatric disorders, including ADHD or schizophrenia with high accuracy (AUC = 0.863 and AUC = 0.853, respectively) [51]. However, the technology is still at an infancy phase and there are many obstacles and limitations to overcome in order to apply it clinically. For example, each algorithm is developed and assigned for each disease and so it is difficult to apply it to other diseases. Moreover, the sample size of each algorithm is too small to apply to public [50].

    Various research efforts are ongoing in order to improve diagnosis, prognosis, and treatment in neuropsychiatry: PGx data and their treatment outcomes have been collected to support data-driven clinical decision-making for the patient. To this end, relations between genetic variation and variable drug responses to psychiatric medications should be well established [50]. The use of AI and machine learning analyses to predict individual-specific responses to psychiatric medications is challenging, but well worth pursuing [50].

    The entry is from 10.3390/genes11121445

    References

    1. World Health Organization. Mental Health. Available online: https://www.who.int/health-topics/mental-health (accessed on 25 November, 2020).
    2. Wainberg, M.L.; Scorza, P.; Shultz, J.M.; Helpman, L.; Mootz, J.J.; Johnson, K.A.; Neria, Y.; Bradford, J.-M.E.; Oquendo, M.A.; Arbuckle, M.R.; et al. Challenges and opportunities in global mental health: A research-to-practice perspective. Curr. Psychiatry Rep. 2017, 19, 28.
    3. Clous, E.; Beerthuizen, K.; Ponsen, K.J.; Luitse, J.; Olff, M.; Goslings, C. Trauma and psychiatric disorders: A systematic review. J. Trauma Acute Care Surg. 2017, 82, 794–801.
    4. Lozupone, M.; Seripa, D.; Stella, E.; La Montagna, M.; Solfrizzi, V.; Quaranta, N.; Veneziani, F.; Cester, A.; Sardone, R.; Bonfiglio, C. Innovative biomarkers in psychiatric disorders: A major clinical challenge in psychiatry. Expert Rev. Proteom. 2017, 14, 809–824.
    5. Eichelbaum, M.; Ingelman-Sundberg, M.; Evans, W.E. Pharmacogenomics and individualized drug therapy. Annu. Rev. Med. 2006, 57, 119–137.
    6. Lauschke, V.M.; Zhou, Y.; Ingelman-Sundberg, M. Novel genetic and epigenetic factors of importance for inter-individual differences in drug disposition, response and toxicity. Pharmacol. Ther. 2019, 197, 122–152.
    7. Sullivan, P.F.; Agrawal, A.; Bulik, C.M.; Andreassen, O.A.; Børglum, A.D.; Breen, G.; Cichon, S.; Edenberg, H.J.; Faraone, S.V.; Gelernter, J.; et al. Psychiatric genomics: An update and an agenda. Am. J. Psychiatry 2018, 175, 15–27.
    8. Hung, S.-I.; Chung, W.-H.; Jee, S.-H.; Chen, W.-C.; Chang, Y.-T.; Lee, W.-R.; Hu, S.-L.; Wu, M.-T.; Chen, G.-S.; Wong, T.-W.; et al. Genetic susceptibility to carbamazepine-induced cutaneous adverse drug reactions. Pharm. Genom. 2006, 16, 297–306.
    9. Dean, L. Carbamazepine Therapy and HLA Genotype. In Medical Genetics Summaries; National Center for Biotechnology Information: Bethesda, MD, USA, 2012.
    10. Fricke-Galindo, I.; LLerena, A.; Jung-Cook, H.; López-López, M. Carbamazepine adverse drug reactions. Expert Rev. Clin. Pharmacol. 2018, 11, 705–718.
    11. Barbarino, J.M.; McGregor, T.L.; Altman, R.B.; Klein, T.E. PharmGKB summary: Very important pharmacogene information for MT-RNR1. Pharm. Genom. 2016, 26, 558.
    12. Rahman, S.; Ecob, R.; Costello, H.; Sweeney, M.G.; Duncan, A.J.; Pearce, K.; Strachan, D.; Forge, A.; Davis, A.; Bitner-Glindzicz, M.; et al. Hearing in 44–45 year olds with m.1555A > G, a genetic mutation predisposing to aminoglycoside-induced deafness: a population based cohort study. BMJ Open 2012, 2, doi:10.1136/bmjopen-2011-000411.
    13. Van der Wouden, C.H.; van Rhenen, M.H.; Jama, W.O.; Ingelman-Sundberg, M.; Lauschke, V.M.; Konta, L.; Schwab, M.; Swen, J.J.; Guchelaar, H.J. Development of the PGx-passport: A panel of actionable germline genetic variants for pre-emptive pharmacogenetic testing. Clin. Pharmacol. Ther. 2019, 106, 866–873.
    14. U.S. Food & Drug Administration. Table of Pharmacogenomic Biomarkers in Drug Labeling. Available online: https://www.fda.gov/drugs/science-and-research-drugs/table-pharmacogenomic-biomarkers-drug-labeling (accessed on 1 November, 2020).
    15. The Clinical Pharmacogenetics Implementation Consortium. Guidelines. Available online: https://web.archive.org/web/20201101003826/ or https://cpicpgx.org/guidelines/ (accessed on 1 November, 2020).
    16. The Canadian Pharmacogenomics Network for Drug Safety (CPNDS). Available online: https://web.archive.org/web/20201125025807/http://cpnds.ubc.ca/ (accessed on 25 November, 2020).
    17. Ross, C.J.; Visscher, H.; Sistonen, J.; Brunham, L.R.; Pussegoda, K.; Loo, T.T.; Rieder, M.J.; Koren, G.; Carleton, B.C.; Hayden, M.R.; et al. The canadian pharmacogenomics network for drug safety: A model for safety pharmacology. Thyroid 2010, 20, 681–687.
    18. Swen, J.J.; Huizinga, T.W.; Gelderblom, H.; De Vries, E.G.; Assendelft, W.J.; Kirchheiner, J.; Guchelaar, H.-J. Translating pharmacogenomics: challenges on the road to the clinic. PLoS Med. 2007, 4, e209, doi:10.1371/journal.pmed.0040209.
    19. Ehmann, F.; Caneva, L.; Prasad, K.; Paulmichl, M.; Maliepaard, M.; Llerena, A.; Ingelman-Sundberg, M.; Papaluca-Amati, M. Pharmacogenomic information in drug labels: European medicines agency perspective. Pharm. J. 2015, 15, 201–210.
    20. Drozda, K.; Pacanowski, M.A.; Grimstein, C.; Zineh, I. Pharmacogenetic labeling of FDA-approved drugs: a regulatory retrospective. JACC Basic Transl. Sci. 2018, 3, 545–549.
    21. The Pharmacogenomics Knowledgebase. Clinical Guideline Annotations. Available online: https://www.pharmgkb.org/guidelineAnnotations (accessed on 1 November, 2020).
    22. Guengerich, F.P. Cytochrome P450 research and the journal of biological chemistry. J. Biol. Chem. 2019, 294, 1671–1680.
    23. Manikandan, P.; Nagini, S. Cytochrome P450 structure, function and clinical significance: A review. Curr. Drug Targets 2018, 19, 38–54.
    24. Pharmacogene Variation Consortium. CYP2D6 Allele Nomenclature. Available online: https://web.archive.org/web/20201031133042/ or https://www.pharmvar.org/gene/CYP2D6 (accessed on 31 October, 2020).
    25. El-Mallakh, R.S.; Roberts, R.J.; El-Mallakh, P.L.; Findlay, L.J.; Reynolds, K.K. Pharmacogenomics in psychiatric practice. Clin. Lab. Med. 2016, 36, 507–523.
    26. Gaedigk, A.; Sangkuhl, K.; Whirl-Carrillo, M.; Klein, T.; Leeder, J.S. Prediction of CYP2D6 phenotype from genotype across world populations. Genet. Med. 2017, 19, 69–76.
    27. LLerena, A.; Naranjo, M.E.G.; Rodrigues-Soares, F.; Penas-LLedo, E.M.; Farinas, H.; Tarazona-Santos, E. Interethnic variability of CYP2D6 alleles and of predicted and measured metabolic phenotypes across world populations. Expert Opin. Drug Metab. Toxicol. 2014, 10, 1569–1583.
    28. Oshiro, C.; Thorn, C.F.; Roden, D.M.; Klein, T.E.; Altman, R.B. KCNH2 pharmacogenomics summary. Pharm. Genom. 2010, 20, 775.
    29. Gaedigk, A.; Gotschall, R.R.; Forbes, N.S.; Simon, S.D.; Kearns, G.L.; Leeder, J.S. Optimization of cytochrome P4502D6 (CYP2D6) phenotype assignment using a genotyping algorithm based on allele frequency data. Pharm. Genom. 1999, 9, 669–682.
    30. Bradford, L.D. CYP2D6 allele frequency in european caucasians, Asians, Africans and their descendants. Pharmacogenomics 2002, 3, 229–243.
    31. Yokota, H.; Tamura, S.; Furuya, H.; Kimura, S.; Watanabe, M.; Kanazawa, I.; Kondo, I.; Gonzalez, F.J. Evidence for a new variant CYP2D6 allele CYP2D6J in a japanese population associated with lower In Vivo rates of sparteine metabolism. Pharmacogenetics 1993, 3, 256–263.
    32. Sistonen, J.; Sajantila, A.; Lao, O.; Corander, J.; Barbujani, G.; Fuselli, S. CYP2D6 worldwide genetic variation shows high frequency of altered activity variants and no continental structure. Pharm. Genom. 2007, 17, 93–101.
    33. Ingelman-Sundberg, M. Genetic polymorphisms of cytochrome P 450 2D6 (CYP2D6): Clinical consequences, evolutionary aspects and functional diversity. Pharm. J. 2005, 5, 6–13.
    34. Ingelman-Sundberg, M.; Sim, S.C.; Gomez, A.; Rodriguez-Antona, C. Influence of cytochrome P450 polymorphisms on drug therapies: pharmacogenetic, pharmacoepigenetic and clinical aspects. Pharmacol. Ther. 2007, 116, 496–526.
    35. Schroth, W.; Hamann, U.; Fasching, P.A.; Dauser, S.; Winter, S.; Eichelbaum, M.; Schwab, M.; Brauch, H. CYP2D6 polymorphisms as predictors of outcome in breast cancer patients treated with tamoxifen: expanded polymorphism coverage improves risk stratification. Clin. Cancer Res. 2010, 16, 4468–4477.
    36. De Vos, A.; Van Der Weide, J.; Loovers, H. Association between CYP2C19* 17 and metabolism of amitriptyline, citalopram and clomipramine in Dutch hospitalized patients. Pharm. J. 2011, 11, 359–367.
    37. Steimer, W.; Zopf, K.; von Amelunxen, S.; Pfeiffer, H.; Bachofer, J.; Popp, J.; Messner, B.; Kissling, W.; Leucht, S. Amitriptyline or not, that is the question: Pharmacogenetic testing of CYP2D6 and CYP2C19 identifies patients with low or high risk for side effects in amitriptyline therapy. Clin. Chem. 2005, 51, 376–385.
    38. Hicks, J.; Swen, J.; Thorn, C.; Sangkuhl, K.; Kharasch, E.; Ellingrod, V.; Skaar, T.; Müller, D.; Gaedigk, A.; Stingl, J.; et al. Clinical pharmacogenetics implementation consortium guideline for CYP2D6 and CYP2C19 genotypes and dosing of tricyclic antidepressants. Clin. Pharmacol. Ther. 2013, 93, 402–408.
    39. Ma, M.K.; Woo, M.H.; Mcleod, H.L. Genetic basis of drug metabolism. Am. J. Health-Syst. Pharm. 2002, 59, 2061–2069.
    40. Weinshilboum, R.; Wang, L. Pharmacogenomics: bench to bedside. Nat. Rev. Drug Discov. 2004, 3, 739–748.
    41. Lazaridis, K.N.; Mcallister, T.M.; Babovic-Vuksanovic, D.; Beck, S.A.; Borad, M.J.; Bryce, A.H.; Chanan-Khan, A.A.; Ferber, M.J.; Fonseca, R.; Johnson, K.J. Implementing individualized medicine into the medical practice. In Proceedings of the American Journal of Medical Genetics Part C: Seminars in Medical Genetics; pp. 15–23.
    42. Evans, W.E.; Relling, M.V. Moving towards individualized medicine with pharmacogenomics. Nature 2004, 429, 464–468.
    43. Collins, F.S.; Varmus, H. A new initiative on precision medicine. New Engl. J. Med. 2015, 372, 793–795.
    44. Iniesta, R.; Stahl, D.; McGuffin, P. Machine learning, statistical learning and the future of biological research in psychiatry. Psychol. Med. 2016, 46, 2455–2465.
    45. Lane, H.-Y.; Tsai, G.E.; Lin, E. Assessing gene-gene interactions in pharmacogenomics. Mol. Diagn. Ther. 2012, 16, 15–27.
    46. Lin, E.; Hwang, Y.; Liang, K.-H.; Chen, E.Y. Pattern-recognition techniques with haplotype analysis in pharmacogenomics. 2007, 8, doi:10.2217/14622416.8.1.75.
    47. Russell, S.J.; Norvig, P. Artificial Intelligence: A Modern Approach. Malaysia; Pearson Education Limited: London, UK, 2016.
    48. Bzdok, D.; Meyer-Lindenberg, A. Machine learning for precision psychiatry: opportunities and challenges. Biol. Psychiatry Cogn. Neurosci. Neuroimaging 2018, 3, 223–230.
    49. Fernandes, B.S.; Williams, L.M.; Steiner, J.; Leboyer, M.; Carvalho, A.F.; Berk, M. The new field of “precision psychiatry”. BMC Med. 2017, 15, 1–7.
    50. Lin, E.; Lin, C.-H.; Lane, H.-Y. Precision psychiatry applications with pharmacogenomics: Artificial intelligence and machine learning approaches. Int. J. Mol. Sci. 2020, 21, 969.
    51. Miotto, R.; Li, L.; Kidd, B.A.; Dudley, J.T. Deep patient: an unsupervised representation to predict the future of patients from the electronic health records. Sci. Rep. 2016, 6, 1–10.
    More