Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Sep 9.
Published in final edited form as: Psychiatr Genet. 2012 Jun;22(3):123–129. doi: 10.1097/YPG.0b013e328353956a

Linkage analysis of alternative anxiety phenotypes in multiply affected panic disorder families

Abby J Fyer 1,2, Ramiro Costa 2, Fatemeh Haghighi 1,2, Mark W Logue 3, James A Knowles 4, Myrna M Weissman 1,2, Susan E Hodge 2,5, Steven P Hamilton 6
PMCID: PMC3767163  NIHMSID: NIHMS370289  PMID: 22525237

Abstract

Background

The choice of phenotype definitions for genetic studies of panic and phobic disorders is complicated by family, twin and neurobiological data indicating both distinct and shared risk factors as well as heterogeneity within categories. We previously reported a genome scan in 120 multiplex panic disorder (PD) families using a phenotype that closely adhered to the DSM IV PD definition. Here we extend this work by conducting exploratory linkage analyses in this same pedigree set using ten additional literature- based panic and phobia-related phenotypes that take into account aspects of these hypothesized complexities.

Methods

Multiply affected families (> 2 individuals with PD) were recruited from clinical and non-clinical sources, evaluated by clinician administered semi-structured interview and subsequent blind consensus best estimate procedure. Each phenotype was analyzed under dominant and recessive models using parametric 2-point (homogeneity and heterogeneity), multipoint, and non-parametric methods. Empirically based permutations were used to estimate model specific and global (across all phenotypes) p-values.

Results

The highest score was a 2-point lod (4.27, global p < 0.08) on chromosome 13 (D13S793, 76cM) for the phenotype “specific or social phobia” under a recessive model and conditions of homogeneity. There was minimal support for linkage to any of the remaining nine phenotypes.

Conclusions

Though interpretation of findings is limited by sample size and the large number of phenotypes and models analyzed these data suggest a region on chromosome 13 as a potential site for further exploration in relation to risk for specific and social phobias.

Keywords: panic, linkage, anxiety, specific phobia, social phobia, phenotype

Introduction

Panic disorder (PD) is a highly familial, genetically complex disorder with heritability estimated at between .3-.5. Family and twin study data indicate shared genetic risk factors among PD and other anxiety disorders, particularly phobias, 3, 4, 5 as well as heterogeneity within the category4, 6, and 7. We previously reported a genome scan of 120 multiplex families affected with PD. These initial analyses used a prospectively defined phenotype that closely adhered to the DSM IV definition of the disorder1, 2. Here we extend this work by conducting linkage analyses of additional panic and phobia-related phenotypes in the same dataset. Our initial phenotypic definitions are consistent with certain aspects of the twin, family study and psychobiological literature, but there are also compelling data supporting alternative approaches.

On the basis of the literature, particularly recent molecular genetic work, we chose ten panic-related phenotypes and conducted separate linkage analyses of each phenotype in our pedigree set. Table 1 summarizes these phenotypes and previous linkage analyses that have used these (or very similar) definitions. For ease of exposition we have classified the ten phenotypes into three broad categories: 1) composite phenotypes (i.e. phenotypes derived from hypotheses of shared genetic risk factors between PD and other disorders); 2) PD subtypes; and 3) phobias (specific and/or social) that segregate in these families with a high enough frequency to enable a separate linkage analysis.

TABLE 1.

Panic disorder related phenotypes with maximum lod score locations in published linkage studies

Phenotype Category Phenotypes analyzed in this report Previous studies using same or similar phenotype1
Investigator Location Max Lod p
Panic related phenotypes from previous PD genome scans Broad Anxiety (any anxiety or somatoform disorder in families identified thru PD proband) Thorgeirsson et al, (2003)2 D9S271 (105.6cM) Lod =4.18 .05 (genome-wide)
PD or Phobias (PD, social, specific or agoraphobia) Kaabi et al (2006)3 (Gelernter pedigrees) D4S413 (158cM) Lod>4 .0006 (empirical)
Early co-morbid anxiety (PD and either PD or other anxiety at <13yrs Smoller et al, (2001)4 D1S1678 (218cM) NPL=2.05 .035
D10S587(147cM) Lod =2.38 --
PD subtypes PD/Agoraphobia Gelernter et al (2001) D3S1279(167M) NPL=2.75 .005
PD/No Agoraphobia NONE
Early onset PD (< 20 yrs) NONE
Childhood onset PD (<13 yrs) NONE
Phobias segregating in PD pedigrees Specific phobia Gelernter et al. (2003) D14S75 (37cM) Lod =3.17 --
Chr 14 (35cM) Zlr =3.93 .00005
Social phobia Gelernter et al. (2004) Chr 16 (62cM) Zlr =3.41 .0003
Chr 16 (71cM) Lod =2.22 ---
Specific or social phobia NONE
Open in a new tab
1

Differences between phenotype as defined in published report and phenotypes used in our analyses are given in the footnotes below

2

Thorgeirsson et al. included as affected: PD, GAD, social or specific phobia, somatoform disorder or subthreshold somatoform pain. We did not have data on the subthreshold somatoform syndrome. In addition we had only 4 cases of somatoform so our criteria or measurement methods most likely differed from theirs.

3

Kaabi et al. used a fuzzy cluster method that gives weighted values to each diagnosis, here we include individuals as affected if they have any one of the three disorders, giving equal weight to each and ignoring comorbidity.

4

Smoller et al. required onset of anxiety in childhood, PD and persistence of anxiety disorder into adulthood. As only lifetime diagnoses were coded, in some cases the latter criterion could not be confirmed. Subjects in our analyses were considered affected if they met the first two criteria.

Below we give a brief rationale for the choice of these phenotypes, organized in terms of these three broad categories. As the multiple testing problem constrains interpretation of these re-analyses, we consider this approach an exploratory (i.e. hypothesis -generating) tool.

Composite phenotypes

These phenotypes are derived from hypotheses of shared genetic risk factors between PD and other disorders. Each of the three is modeled on a phenotype for which suggestive evidence of linkage was found in a recent PD genome scan (Table 1)8, 9, 10. The broadest approach is taken by Thorgeirsson and colleagues,8 who identified pedigrees through an individual with PD, but defined affectedness in relatives to include PD, social, specific or agoraphobia, generalized anxiety disorder (GAD), or somatoform pain. The narrower “PD or phobias” category was suggested by both twin study findings of overlap in the genetic risk factors for PD, social, specific and agoraphobia11, 5 and a recent collaborative re-analysis of 19 pedigrees12, 9. This analysis, incorporating a “fuzzy clustering” approach, facilitates a fine-grained assessment of the interrelationships among the disorders using weighted likelihoods of affectedness of each family member for each of the three disorders. Early comorbid anxiety and the more restrictive phenotype (Diathesis or D-Type PD) tested here were defined by Smoller and colleagues, 10 based on results of longitudinal and family studies of the childhood temperament behavioral inhibition (BI). This phenotype includes individuals who have PD, and pre-pubertal onset of either the PD or another anxiety disorder.

Panic disorder subtypes

Three PD subtypes (PD with agoraphobia [PD+A]; PD without agoraphobia [PD-A]; early-onset PD) have been the subject of considerable study, including work suggesting heritable differences. For example, Kendler et al4 found both specific and shared (with PD and GAD) genetic contributions to agoraphobia; Noyes et al6 found equal rates of panic attacks but increased rates of panic disorder with agoraphobia (PDA) among the relatives of probands with PDA when compared with relatives of PD probands. Early onset of PD (< 20 years), has been associated with increased severity, duration, and psychiatric comorbidity, and significantly higher rates of PD have been found in families of probands with early-onset PD when compared to families of probands with late-onset PD13, 14, 15. However, there are no molecular genetic data addressing the genetic basis of these observations, and the one family study that examined onset in relatives found no relationship between age of onset in the proband (< 20 years) and age of onset in an affected relative (i.e. early-onset probands did not have early-onset relatives)13. We chose to test two definitions of early-onset PD, the < 20 years and < 13 (i.e. pre-pubertal PD).

Specific and social phobia

Twin studies indicate a moderate genetic contribution to risk for both social and specific phobia, with heritability usually in the 0.3-0.6 range16, 17, 18, 19, 20. Linkage analyses in pedigree sets collected specifically to study phobias have not been reported. However, Gelernter and colleagues21, 22 conducted linkage analyses in a set of panic disorder pedigrees, successively using each social and specific phobia (rather than panic disorder) as the “affected” phenotype (Table 1). We use the same strategy here in our collection of pedigrees collected for PD. In addition, as twin study data indicate some overlap between genetic risk factors for social and specific phobia, we include an additional composite phenotype: “social or specific phobia”23, 5.

Methods

Sample

Recruitment and evaluation of this pedigree sample were previously described1, 2. The study was approved by the New York State Psychiatric Institute Institutional Review Board, and all subjects gave informed consent before participating. Briefly, we recruited families from clinical and non-clinical sources who at screening appeared to have at least three affected members willing to be interviewed and to give a blood sample for DNA. All available subjects were interviewed by specially trained clinicians using the SADS L-Anxiety Version (SADS-LA) 24, 25 which provides lifetime DSM III-R or IV diagnoses for most psychiatric disorders, as well as more detailed information about anxiety disorders. Interviewed individuals provided family history information about those relatives thought to have anxiety disorders. Interviewers wrote detailed narratives describing the development of psychiatric symptoms in the context of subject’s social history26. Depending on phenotype prevalence, each analysis included between 51 and 120 pedigrees (Table 2).

Table 2.

Linkage analyses of panic-related phenotypes: maximum lod scores

Phenotype Category Phenotype Current study sample size Current Study: Maximum Lod3
Number of Families Individuals Genotyped (Affected/All) Chrom Marker (cM) Model Lod Score1
Composite PD-related phenotypes based on previous scans2 Broad anxiety (PD, GAD, Soc, Spec, somatoform) 120 869/992 2 D2S125 (261) Dominant Sex-dif, Hetlod 2.64
PD or phobias (PD, specific social, or agoraphobia) 3 120 861/992 2 D2S125 (261) Dominant Sex-dif, Hetlod 2.52
Early comorbid anxiety (PD & anxiety disorder <13 years) 120 247992 2 D2S1788 (56) Recessive Sex dif, Hetlod 1.96
Panic Disorder Subtypes PD no Agoraphobia 112 379/916 16 N/A (141) Dominant MP, Hetlod 2.76†
PD with Agoraphobia 92 271/824 7 D7S821 (109) Recessive Sex-dif, Hetlod 2.67
Early onset PD (≤20) 91 213/767 4 D4S2361 (93) Dominant 2 point, Hetlod 2.29
Early onset PD (≤13) 51 81/734 20 D20S480 (80) Recessive 2point, Homlod 2.69†
Anxiety Disorders Segregating in PD pedigrees Specific Phobia 69 247/644 13 D13S793 (76) Recessive 2point, Homlod 2.77*#
Social Phobia 61 213/609 7 D7S1799 (114) Dominant 2pint, Hetlod 2.28
Specific or Social Phobia 93 386/816 13 D13S793 (76) Recessive 2point, Homlod 4.27**Ψ
Open in a new tab
1

Significance: Model specific p values (see text, data analysis)

†

p=0.1;

*

p≤.05;

**

p≤.01.

“Global” p values:

Ψ

p=.076;

#

p=0.9

2

Modified respectively from Thorgeirsson et al, 2004; Kaabi et al, 2006; and Smoller et al, 2001.

3

Abbreviations: Chrom = chromosome; cM=centimorgans, Hetlod = lod calculated under assumption of heterogeneity; Homlod = lod calculated under assumption of homogeneity; MP= multipoint, N/A = not apply; sex-dif = calculated will allowance for gender distinction in recombination fraction.

Genotyping

Microsatellite genotyping was conducted at the Center for Inherited Disease Research (CIDR) following the standard CIDR protocol, with a marker set including 384 simple tandem repeats at an average spacing of 9cM, with no gaps greater than 20cM. Quality control procedures were previously described.2

Diagnosis

PD diagnoses and age at onset were arrived at through a best estimate process1 in which all available materials were independently reviewed by at least two senior clinicians and classified in one of six levels of panic disorder affectedness (definite, probable, possible, “any panic”, unaffected, and unknown). We further defined three PD thresholds: Broad (definite, probable, possible, “any panic”), Intermediate (excludes “any panic”); and Narrow (definite or probable). All other diagnoses used to define the various phenotypes (e.g. social and specific phobia, agoraphobia, GAD) are the DSM III R or DSM IV diagnoses made by the clinical interviewer using the SADS-LA. Narratives and interview forms were reviewed by a PhD-level experienced research clinician for diagnostic validity, but a formal best estimate was not carried out for these diagnoses.

Classification of individuals and families for linkage analyses

Phenotype definitions and criteria for including families are described in this section. In all analyses: 1) those individuals who did not meet the specific affected and unaffected criteria were classified as “unknown”; and 2) all 120 families were used unless otherwise specified. This set of 120 families is the same as was reported in our 2006 using article using a PD phenotype that closely adhered to the DSM IV definition.

Composite PD-related phenotypes

Broad anxiety: Individuals were included as “affected” if they met criteria for any of these anxiety disorders: Broad PD with or without agoraphobia, GAD, or social or specific phobia. “Unaffecteds” met our “PD unaffected” criteria and did not have any of the above listed anxiety disorders. PD or phobias: Affected was defined as having Broad PD or specific or social phobia, and unaffected, as meeting our “PD unaffected” criteria and not having social or specific phobia. Early comorbid anxiety: Affected was defined as: 1) Broad PD at or before age 13 years or 2) another anxiety disorder at or before age 13 and PD at any age. Individuals without PD were classified as unaffected.

PD subtypes

For each analysis, families with at least one person who met the affected criteria were included; unaffected was defined as our best estimate unaffected category. Panic disorder without agoraphobia (PD-A): Affected individuals met criteria for definite, probable or possible PD but not for agoraphobia. Panic disorder with agoraphobia (PD+A): Affected individuals met criteria for both any level of PD and agoraphobia. Early onset PD: We analyzed two definitions: ≤ 20 and ≤ 13 years. In each case affected individuals met the Broad PD criteria by the given age.

Phobic disorders

These three analyses followed the same format. Individuals were considered affected if they met the DSM IIIR or IV criteria for the phobic disorder studied; families with at least two persons with the phobia diagnosis were included in the analysis; and all persons without the diagnosis were classified as unaffected. Individuals were considered affected if they met criteria for social phobia in the first analysis, for specific phobia in the second analysis, and for either social or specific phobia in the third analysis.

Linkage analysis

Statistical procedures were the same as those in our primary report2. Each phenotype was analyzed under two genetic models (dominant and recessive), using both parametric (2-point, multipoint) and non-parametric methods. We performed 2-point parametric analyses twice, first using sex-averaged recombination fractions, then allowing for gender-specific recombination fractions. FASTLINK27 and ANALYZE28 were used for the 2-point analyses and GENEHUNTER29, for the multipoint parametric and nonparametric calculations.

Although we consider these analyses to be exploratory, it nevertheless seemed of interest to have a systematic evaluation of the relative significance of the linkage findings, though to do so required taking into account the large number of models and phenotypes examined. To this end, and given the computational intensity, we adopted a two-step permutation based approach. First we estimated empirical p-values within each phenotype using 100 randomized replicates. Then, for our best-supported scores, we estimated phenotype-specific and “global” (maximized across all models and phenotypes) p-values in 10,000 replicates created through bootstrap re-sampling. The randomized replicates of the data were generated by random assignment of genotypes to the founders within each family, while conditioning on the observed allele frequency and inter-marker distances in the data, and then matching individuals within families according to the observed pedigree structure using the program SIMULATE30. In this way, 100 randomized replicates of the data were generated and then analyzed for each of the genetic markers and models (dominant, recessive) considered in this study.2 Model-specific p-values were estimated within each phenotype maximizing across genetic models and markers. This estimate is done by recording the number of times a maximum lod score from a replicate exceeds the maximum lod score from that of the observed data, and therefore corrects for the testing of multiple marker loci examined. Lod scores that achieved model specific p < 0.05 in this analysis were further evaluated as follows. First, 100 of the original replicates were used for bootstrap re-sampling 10,000 times. This new sample was then used to estimate both the model-specific and the global empirical p-values at the markers for which lods with p < 0.05 were achieved in the first stage analysis. These analyses were done using the same genetic models (i.e., in this case 2-point lods, under dominant and recessive models, but assuming homogeneity, see Table 2 below) that were used in the first-stage analysis of that marker. The global p-value was estimated by maximizing the lod scores across all genotypic-phenotypic models for each replicate, and then recording the number of times these maximized scores exceed the observed model-specific lod score per replicate. The global p-value accounts for all markers and models tested. Thus it accurately captures the experimental design of this study and provides a reliable measure for interpreting the findings.

Results

Table 2 gives the maximum lod scores and the number of families/subjects included in the analysis for each of our ten phenotypes. The maximum lod scores for each of the ten phenotypes under any model are shown in Table 2. The highest score (lod = 4.27) was a 2-point lod on chromosome 13 (D13S793, 76cM) for the phenotype “specific or social phobia” under a recessive model and conditions of homogeneity. For that analysis, a model-specific p-value was estimated at 0.01 using our initial 100 replicates. In the second permutation analysis, using the 10,000 bootstrapped replicates, we estimated the model-specific p-value at 0.002 and the global p < 0.08. However, multipoint and NPL scores for “specific or social phobia” under the same model were modest (1.13 and 1.05 respectively). At this same location and under the same genetic model, the specific phobia phenotype on its own also achieved a model-specific lod with p-value less than 0.03 in the initial 100 replicate analyses. However, in the larger bootstrapped permutation sample the model-specific p was higher (0.07), and the global p = 0.846. The only other phenotype that achieved a lod score with accompanying p-values less than 0.10 was early-onset PD defined as onset before age 13 years. For this phenotype a 2-point mode-specific lod of 2.69 (p = 0.06) was found using the 100 replicates on chromosome 20 at marker D20S480, under a recessive model and conditions of homogeneity. The corresponding model-specific and global p-values in the 10,000 replicate analyses under the same genetic model were 0.07 and 0.90 respectively. We did not replicate results of previous scans. For each phenotype, maximum lod scores were < 1.0 (data not shown) at the highest scoring locations in this earlier work (Table 1).

Discussion

These exploratory re-analyses provide suggestive evidence for a susceptibility locus for a “social or specific phobia” phenotype on chromosome 13q within families also affected by PD. In this sample there was minimal support for linkage to any of the remaining nine phenotypes.

Though we consider the “specific or social phobia” findings on chromosome 13q of interest for further prospective study, we also emphasize their significant limitations. These include: moderate sample size; the large number of phenotypic, genetic and statistical models analyzed; and the lack of convergent support from our nonparametric and multipoint analyses; though, the latter may simply represent a loss of power due to GENEHUNTER’s trimming of samples (a necessity in analysis of complex pedigrees) rather than divergent evidence from multipoint and 2-point methods. To our knowledge the only previous molecular genetic studies of the specific or social phobia phenotypes are the reports by Gelernter and colleagues, 21, 22 on which we modeled our phobia analyses. As shown in Table 1, the regions of their maximum lod scores were on chromosomes 14 and 16 respectively. Lod scores at these sites in our data set were < 1.0 for all three phobia phenotypes under all models. Gelernter and colleagues21, 22 did not report data on the combined phobia phenotype.

The 10 Mb interval surrounding our peak (D13S793 at 76cM) includes over 30 genes. Though none of these genes has been previously associated with PD, several are intriguing candidates, as they are involved in aspects of neural development or function (e.g., FGF14, NALCN). However, as there is little knowledge of the etiology of phobias at the molecular level 32 it is difficult to construct specific hypotheses. In addition, though the genes that investigators discover for complex diseases are not necessarily those that anyone had predicted ahead of time would be involved in those diseases33, we are not aware of data supporting any mechanism that would particularly link phobias to any of these processes. Interestingly, a small genome-wide association study of PD initially reported an association to a DNA variant just 1.4Mb from D13S793 (p = 3×10-7), but the finding was not replicated in an enlarged sample.34, 35

Of note, our previous study reported linkage between this same region on chromosome 13 (76cM, D13S793, 2 point HLOD = 3.57) and a potentially pleiotropic PD syndrome31, 7 characterized by aggregation of PD and several medical disorders (renal/bladder problems, thyroid disease, migraines and other serious headaches, mitral valve prolapse) within a subset (N = 60) of our original120-family PD pedigree set. Given the available data it is not possible to determine whether this convergence reflects pleiotropy (i.e. phobias are also part of this previously identified syndrome) or two independent findings within the same region. The maximum scores for the “syndrome” and our phobia phenotype did occur under different genetic models (dominant and recessive, respectively). The 2-point lod score for “social or specific phobia” at D13S793 under a dominant model was 1.1. In addition, when we used a combined phenotype (affected = “syndrome” or “social or specific phobia”) in a post hoc analysis within the 60 original syndrome families7, the 2-point lod scores at D13S793 were considerably lower than those found using the original syndrome phenotype and did not provide substantial evidence for linkage (recessive model, homogeneity 0.45, heterogeneity 0.80; dominant model, homogeneity 0.83, heterogeneity 1.1). However, though these observations are consistent with independence they do not address the possibility of genetic heterogeneity within our specific/social phobia cases, or indirect influence of multiple interacting genes. Therefore pleiotropy cannot be ruled out.

The possibility of a genetic variant that confers risk for either social or specific phobia (e.g. “phobia proneness”) is consistent with previous genetic epidemiological data, though in some studies the best-fitting models suggest that social phobia has a closer relationship to PD or generalized anxiety than to specific phobia. For example, structural modeling analyses in the Virginia Twin Registry23, 5 indicated two major additive genetic factors that contribute to anxiety disorders. Of these, one contributes roughly half the liability for situational and animal phobias and 15% of the liability for social phobia. The second factor contributes about 30% of the liability for social phobia, but only 10% for each type of specific phobia. Similarly, a Norwegian population-based twin/family study of irrational fears indicated that a genetic factor common to all irrational fears contributed 20% of the liability for animal and situational fears, but only 1% of liability for social fears36.

Given the current experience in psychiatric genetics we do not find our failure to replicate previous work surprising.37, 38 The most likely explanation is moderate sample size. Although our dataset is larger than most of those used to generate the original signals, none of the panic-related linkage samples described here would be expected to have sufficient power to detect linkage for genes contributing the small increases in liability that seem to characterize complex disorders.39 In addition to our moderate sample sizes and significance levels (with their accompanying risk of false positives in the context of a study of genetically complex disorders), we also note: cross-study differences in diagnostic conventions and populations (e.g. Thorgierssen’s population based vs our volunteer sample8); inaccuracies inherent in our attempt to retrospectively match diagnostic criteria for certain phenotypes (e.g. D type, broad anxiety) and the absence of analyses considering comorbid conditions (e.g. major depression, substance and alcohol abuse). In some cases 10,12 the analytic strategies and model parameters we used were almost identical to those in the previous reports or differed only in ways 21 unlikely to have a large effect on outcome40. However, in the case of the studies by Kaabi et al 9 and Thorgierssen8 there were significant analytic differences that must be taken into account in evaluating our results.

In summary, we conducted exploratory linkage analyses of ten panic/phobia related phenotypes suggested by the literature in a previously analyzed panic pedigree set. We did not replicate any of the findings of previous investigators using similar phenotypes in panic disorder pedigrees. The analysis of a combined phobia phenotype (specific or social phobia) suggests a region on chromosome 13 as a potential site for further exploration in relation to risk for these disorders.

Acknowledgments

This work was supported in part by NIMH Grants MH28274 (to MMW), MH37592 (to AJF), MH30906 (to DFK), MH48858 (to SEH), HG002915 (to FH), and MH074118 (to FH) K01 MH076100 (to MWL). Genotyping services were provided to JAK by the Center for Inherited Disease Research (CIDR). CIDR is fully funded through a federal contract from the National Institutes of Health to Johns Hopkins University; contract number N01-HG-65403. We thank the families who participated for making this work possible.

Footnotes

Conflict of Interest Disclosures

Drs. Fyer, Costa, Logue, Haghighi, Hamilton and Hodge reported no biomedical financial interests or potential conflicts of interest. Dr. Knowles is on a Scientific Advisory Board for Life Technologies, Inc. and on the technical advisory board of SoftGenetics, Inc. In the past two years, Dr. Weissman received funding from the National Institute of Mental Health (NIMH), the National Institute on Drug Abuse (NIDA), the National Alliance for Research on Schizophrenia and Depression (NARSAD), the Sackler Foundation, the Templeton Foundation and the Interstitial Cystitis Association; and receives royalties from the Oxford University Press, Perseus Press, the American Psychiatric Association Press, and MultiHealth Systems.

References

  1. Fyer AJ, Weissman MM. Genetic linkage study of panic: Clinical methodology and description of pedigrees. American Journal of Medical Genetics. 1999;88(2):173–181. [PubMed] [Google Scholar]
  2. Fyer AJ, Hamilton SP, Durner M, Haghighi F, Heiman GA, Costa R, et al. A third-pass genome scan in panic disorder: Evidence for multiple susceptibility loci. Biological Psychiatry. 2006;60(4):388–401. doi: 10.1016/j.biopsych.2006.04.018. [DOI] [PubMed] [Google Scholar]
  3. Goldstein RB, Weissman MM, Adams PB, Horwath E, Lish JD, Charney D, et al. Psychiatric disorders in relatives of probands with panic disorder and/or major depression. Archives of General Psychiatry. 1994;51(5):383–394. doi: 10.1001/archpsyc.1994.03950050043005. [DOI] [PubMed] [Google Scholar]
  4. Kendler KS, Neale MC, Kessler RC, Heath AC, Eaves LJ. Panic disorder in women: A population-based twin study. Psychological Medicine. 1993;23:397–406. doi: 10.1017/s003329170002849x. [DOI] [PubMed] [Google Scholar]
  5. Hettema JM, Prescott CA, Myers JM, Neale MC, Kendler KS. The structure of genetic and environmental risk factors for anxiety disorders in men and women. Archives of General Psychiatry. 2005;62(2):182–189. doi: 10.1001/archpsyc.62.2.182. [DOI] [PubMed] [Google Scholar]
  6. Noyes R, Jr, Crowe RR, Harris EL, Hamra BJ, McChesney CM, Chaudhry DR. Relationship between panic disorder and agoraphobia: A family study. Archives of General Psychiatry. 1986;43(3):227–232. doi: 10.1001/archpsyc.1986.01800030037004. [DOI] [PubMed] [Google Scholar]
  7. 7.Hamilton SP, Fyer AJ, Durner M, Heiman GA, Baisre de Leon A, Hodge SE, et al. Further genetic evidence for a panic disorder syndrome mapping to chromosome 13q. Proceedings of the National Academy of Sciences of the United States of America. 2003;100(5):2550–2555. doi: 10.1073/pnas.0335669100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. 8.Thorgeirsson TE, Oskarsson H, Desnica N, Kostic JP, Stefansson JG, Kolbeinsson H, et al. Anxiety with panic disorder linked to chromosome 9q in iceland. The American Journal of Human Genetics. 2003;72(5):1221–1230. doi: 10.1086/375141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. 9.Kaabi B, Gelernter J, Woods SW, Goddard A, Page GP, Elston RC. Genome scan for loci predisposing to anxiety disorders using a novel multivariate approach: Strong evidence for a chromosome 4 risk locus. The American Journal of Human Genetics. 2006;78(4):543–553. doi: 10.1086/501072. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. 10.Smoller JW, Acierno JS, Jr, Rosenbaum JF, Biederman J, Pollack MH, Meminger S, Pava JA, Chadwick LH, White C, Bulzacchelli M, Slaugenhaupt SA. Targeted genome screen of panic disorder and anxiety disorder proneness using homology to murine QTL regions. Am J Med Genet. 2001;105(2):195–206. doi: 10.1002/ajmg.1209. [DOI] [PubMed] [Google Scholar]
  11. 11.Kendler KS, Walters EE, Truett KR, Heath AC, Neale MC, Martin NG, et al. A twin-family study of self-report symptoms of panic-phobia and somatization. Behavior Genetics. 1995;25(6):499–515. doi: 10.1007/BF02327574. [DOI] [PubMed] [Google Scholar]
  12. 12.Gelernter J, Bonvicini K, Page G, Woods SW, Goddard AW, Kruger S, et al. Linkage genome scan for loci predisposing to panic disorder or agoraphobia. American Journal of Medical Genetics. 2001;105(6):548–557. doi: 10.1002/ajmg.1496. [DOI] [PubMed] [Google Scholar]
  13. 13.Goldstein RB, Wickramaratne PJ, Horwath E, Weissman MM. Familial aggregation and phenomenology of ‘early’-onset (at or before age 20 years) panic disorder. Archives of General Psychiatry. 1997;54(3):271–278. doi: 10.1001/archpsyc.1997.01830150097014. [DOI] [PubMed] [Google Scholar]
  14. 14.Venturello S, Barzega G, Maina G, Bogetto F. Premorbid conditions and precipitating events in early-onset panic disorder. Comprehensive Psychiatry. 2002;43(1):28–36. doi: 10.1053/comp.2002.29844. [DOI] [PubMed] [Google Scholar]
  15. 15.Seguí J, Márquez M, García L, Canet J, Salvador-Carulla L, Ortiz M. Differential clinical features of early-onset panic disorder. Journal of Affective Disorders. 1999;54(1-2):109–117. doi: 10.1016/s0165-0327(98)00148-7. [DOI] [PubMed] [Google Scholar]
  16. 16.Rose RJ, Ditto WB. A developmental-genetic analysis of common fears from early adolescence to early adulthood. Child Development. 1983;54(2):361–368. [PubMed] [Google Scholar]
  17. 17.Phillips K, Fulker DW, Rose RJ, Eaves LJ. Path analysis of seven fear factors in adult twin and sibling pairs and their parents. Genetic Epidemiology. 1987;4(5):345–355. doi: 10.1002/gepi.1370040504. [DOI] [PubMed] [Google Scholar]
  18. 18.Torgersen S. The nature and origin of common phobic fears. Br J Psychiatry. 1979;134:343–51. doi: 10.1192/bjp.134.4.343. [DOI] [PubMed] [Google Scholar]
  19. 19.Skre I, Onstad S, Torgersen S, Lygren S, Kringlen E. The heritability of common phobic fear: A twin study of a clinical sample. Journal of Anxiety Disorders. 2000;14(6):549–562. doi: 10.1016/s0887-6185(00)00049-9. [DOI] [PubMed] [Google Scholar]
  20. 20.Kendler KS, Karkowski LM, Prescott CA. Fears and phobias: Reliability and heritability. Psychological Medicine. 1999;29(3):539–53. doi: 10.1017/s0033291799008429. [DOI] [PubMed] [Google Scholar]
  21. 21.Gelernter J, Page GP, Bonvicini K, Woods SW, Pauls DL, Kruger S. A chromosome 14 risk locus for simple phobia: Results from a genomewide linkage scan. Molecular Psychiatry. 2003;8(1):71–82. doi: 10.1038/sj.mp.4001224. [DOI] [PubMed] [Google Scholar]
  22. 22.Gelernter J, Page GP, Stein MB, Woods SW. Genome-wide linkage scan for loci predisposing to social phobia: Evidence for a chromosome 16 risk locus. American Journal of Psychiatry. 2004;161(1):59–66. doi: 10.1176/appi.ajp.161.1.59. [DOI] [PubMed] [Google Scholar]
  23. 23.Kendler KS, Neale MC, Kessler RC, Heath AC, Eaves LJ. The genetic epidemiology of phobias in women: The interrelationship of agoraphobia, social phobia, situational phobia, and simple phobia. Archives of General Psychiatry. 1992;49(4):273–281. doi: 10.1001/archpsyc.1992.01820040025003. [DOI] [PubMed] [Google Scholar]
  24. 24.Fyer AJ, Endicott J, Manuzza S, Klein DF. Schedule for affective disorders and schizophrenia-lifetime version, modified for the study of anxiety disorders, 1985. updated for DSM-IV (SADS-LA-IV) 1995 doi: 10.1016/0022-3956(86)90034-8. [DOI] [PubMed] [Google Scholar]
  25. 25.Mannuzza S, Fyer AJ, Klein DF, Endicott J. Schedule for affective disorders and Schizophrenia—Lifetime version modified for the study of anxiety disorders (SADS-LA): Rationale and conceptual development. Journal of Psychiatric Research. 1986;20(4):317–325. doi: 10.1016/0022-3956(86)90034-8. [DOI] [PubMed] [Google Scholar]
  26. 26.Mannuzza S, Fyer AJ, Klein DF. Assessing psychopathology. Int J Methods Psychiatric Res. 1993;3(72):71–72.79. [Google Scholar]
  27. 27.Becker A, Geiger D, Schäffer AA. Automatic selection of loop breakers for genetic linkage analysis. Hum Hered. 1998 Jan-Feb;48(1):49–60. doi: 10.1159/000022781. [DOI] [PubMed] [Google Scholar]
  28. 28.Hiekkalinna T, Terwilliger JD, Sammalisto S, Peltonen L, Perola M. AUTOGSCAN: powerful tools for automated genome-wide linkage and linkage disequilibrium analysis. Twin Res Hum Genet. 2005 Feb;8(1):16–21. doi: 10.1375/1832427053435382. [DOI] [PubMed] [Google Scholar]
  29. 29.Kruglyak L, Daly MJ, Reeve-Daly MP, Lander ES. Parametric and nonparametric linkage analysis: a unified multipoint approach. Am J Hum Genet. 1996 Jun;58(6):1347–63. [PMC free article] [PubMed] [Google Scholar]
  30. 30.Terwilliger JD, Speer M, Ott J. Chromosome-based method for rapid computer simulation in human genetic linkage analysis. Genetic Epidemiology. 1993;10(4):217–224. doi: 10.1002/gepi.1370100402. [DOI] [PubMed] [Google Scholar]
  31. 31.Weissman MM, Fyer AJ, Haghighi F, Heiman G, Deng Z, Hen R, et al. Potential panic disorder syndrome: Clinical and genetic linkage evidence. American Journal of Medical Genetics. 2000;96(1):24–35. doi: 10.1002/(sici)1096-8628(20000207)96:1<24::aid-ajmg7>3.0.co;2-e. [DOI] [PubMed] [Google Scholar]
  32. 32.Rosen JB, Pagani JH, Rolla KL, Davis C. Analysis of behavioral constraints and the neuroanatomy of fear to the predator odor trimethylthiazoline: a model for animal phobias. Neurosci Biobehav Rev. 2008;32(7):1267–76. doi: 10.1016/j.neubiorev.2008.05.006. [DOI] [PubMed] [Google Scholar]
  33. 33.Greenberg DA, Durner M, Keddache M, Shinnar S, Resor SR, Moshe SL, et al. Reproducibility and complications in gene searches: linkage on chromosome 6, heterogeneity, association, and maternal inheritance in juvenile myoclonic epilepsy. Am J Hum Genet. 2000 Feb;66(2):508–16. doi: 10.1086/302763. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. 34.Otowa T, Yoshida E, Sugaya N, Yasuda S, Nishimura Y, Inoue K, et al. Genome-wide association study of panic disorder in the Japanese population. J Hum Genet. 2009 Feb;54(2):122–6. doi: 10.1038/jhg.2008.17. [DOI] [PubMed] [Google Scholar]
  35. 35.Otowa T, Tanii H, Sugaya N, Yoshida E, Inoue K, Yasuda S, et al. Replication of a genome-wide association study of panic disorder in a Japanese population. J Hum Genet. 2009 Dec 4; doi: 10.1038/jhg.2008.17. [DOI] [PubMed] [Google Scholar]
  36. 36.Sundet JM, Skre I, Okkenhaug JJ, Tambs K. Genetic and environmental causes of the interrelationships between self-reported fears. A study of a non-clinical sample of norwegian identical twins and their families. Scandinavian Journal of Psychology. 2003;44(2):97–106. doi: 10.1111/1467-9450.00326. [DOI] [PubMed] [Google Scholar]
  37. 37.Smoller JW, Block SR. Young MM Genetics of anxiety disorders: the complex road from DSM to DNA Depress Anxiety. 2009;26(11):965–75. doi: 10.1002/da.20623. [DOI] [PubMed] [Google Scholar]
  38. 38.Gorroochurn P, Hodge SE, Heiman GA, Durner M, Greenberg DA. Non-replication in association studies: “Pseudo-failures” to replicate? Genetics in Medicine. 2007;9:325–331. doi: 10.1097/gim.0b013e3180676d79. [DOI] [PubMed] [Google Scholar]
  39. 39.Manolio TA, Collins FS, Cox NJ, Goldstein DB, Hindorff LA, Hunter DJ, et al. Finding the missing heritability of complex diseases. Nature. 2009 Oct 8;461(7265):747–53. doi: 10.1038/nature08494. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. 40.Hodge SE, Spence MA, Drandall BF, Sparkes RS, Sparkes MC, Crist M, Tideman S. Huntington Disease: Linkage analysis with age-of-onset corrections. American Journal of Medical Genetics. 1980;5:247–254. doi: 10.1002/ajmg.1320050305. [DOI] [PubMed] [Google Scholar]

RESOURCES