Clinical Genomic Database

Benjamin D. Solomona,1, Anh-Dao Nguyenb, Kelly A. Beara,c, and Tyra G. Wolfsbergb

a
Medical Genetics Branch and bGenome Technology Branch, National Human Genome Research Institute, National Institutes of Health, Bethesda, MD 20892;
and cDepartment of Pediatrics, Walter Reed National Military Medical Center, Bethesda, MD 20889

Edited by C. Thomas Caskey, Baylor College of Medicine, Houston, TX, and approved May 2, 2013 (received for review February 19, 2013)

Technological advances have greatly increased the availability of variant, and associations of the gene or specific variant with known
human genomic sequencing. However, the capacity to analyze health conditions. After this level of curation, however, a key ob-
genomic data in a clinically meaningful way lags behind the ability stacle arises when trying to determine which detected variants may
to generate such data. To help address this obstacle, we reviewed warrant further follow-up, including potential clinical interventions,
all conditions with genetic causes and constructed the Clinical or would otherwise alter patient-based care. Typically, clinically
Genomic Database (CGD) (http://research.nhgri.nih.gov/CGD/), a oriented analysis involves an approach in which detected potentially
searchable, freely Web-accessible database of conditions based pathogenic variants affecting known disease-associated loci are
on the clinical utility of genetic diagnosis and the availability of largely individually queried to determine their clinical applicability
specific medical interventions. The CGD currently includes a total of (9–13).
2,616 genes organized clinically by affected organ systems and To help address this problem, we manually investigated all
interventions (including preventive measures, disease surveillance, conditions with known genetic causes. We constructed a data-
and medical or surgical interventions) that could be reasonably base focusing on genetic data as relates to the availability of
warranted by the identification of pathogenic mutations. To aid condition-specific interventions and how finding a pathogenic
independent analysis and optimize new data incorporation, the mutation would be anticipated to affect medical care. The cur-
CGD also includes all genetic conditions for which genetic knowl- rent goal of this project is to disseminate the database to solicit
edge may affect the selection of supportive care, informed medical content-oriented input, related to both clinical and molecular
decision-making, prognostic considerations, reproductive deci- aspects of the database, from experts in individual genes and
sions, and allow avoidance of unnecessary testing, but for which conditions. Eventually, this database may be used to aid in the
specific interventions are not otherwise currently available. For efficient analysis of individual genomes for clinically significant
each entry, the CGD includes the gene symbol, conditions, allelic health information. The Clinical Genomic Database (CGD) is
conditions, clinical categorization (for both manifestations and freely available at: http://research.nhgri.nih.gov/CGD.
interventions), mode of inheritance, affected age group, descrip-
tion of interventions/rationale, links to other complementary data- Results
bases, including databases of variants and presumed pathogenic At the time of writing, (April 2013), the CGD includes 2,616 genes
mutations, and links to PubMed references (>20,000). The CGD will in which mutations are known to cause human disease or have
be regularly maintained and updated to keep pace with scientific clinically significant pharmacogenomic implications. For 1,333 of
discovery. Further content-based expert opinions are actively soli- these genes, medical interventions meeting the described criteria
cited. Eventually, the CGD may assist the rapid curation of individ- are available (Materials and Methods). The CGD includes an ad-
ual genomes as part of active medical care. ditional 1,283 genes for which these types of clinical interventions
are not yet available based on current medical knowledge, but in
genome sequencing | genomic medicine | whole-genome sequencing which mutations may nonetheless be clinically relevant. Knowl-
edge of mutations resulting in one of this latter group of conditions

A s a result of new technologies that allow efficient and af-

fordable high-throughput sequencing, genomic sequencing
is becoming increasingly prevalent in both research and clinical
may thus be beneficial for a variety of reasons. These reasons in-
clude an enhanced ability to select optimal supportive care, make
more fully informed medical choices, consider questions related to
arenas. Currently, this type of sequencing commonly includes disease prognosis, make reproductive decisions, and avoid lengthy,
exome sequencing, sometimes referred to as “whole-exome se- expensive, and potentially risky “diagnostic odysseys.”
quencing.” In exome sequencing, the protein-coding portions of The Web interface to the CGD allows searching by gene or
known genes—comprised of ∼1–2% of the 6 billion bases in the condition, as well as browsing by clinical categories (for both

GENETICS
diploid genome, depending on the platform used—are sequenced manifestations and interventions). The CGD can be queried using
(reviewed in ref. 1). Whole-genome sequencing, which addition- single or multiple search terms, including large files of gene names
ally includes introns and gene regulatory regions, as well as the or terms. For each entry, the database includes the gene symbol,
rest of the genome, is anticipated to become more widely used as conditions, allelic conditions, clinical categorization (by manifes-
methodologies evolve to allow decreased cost and to meet in- tation and intervention categories), mode of inheritance, age cat-
formatics challenges. Whole-genome sequencing may supplant egory (pediatric or adult) in which interventions are indicated
exome sequencing in the relatively near future. based on descriptions in the medical literature, general descrip-
To date, the most impressive applications of human genomic tions of the interventions/rationale, and individually linked refer-
sequencing (we refer here to both exome and genome sequencing ences (>20,000). See Table 1 for a summary of the categories
as “genomic sequencing”) have been the detection of the genetic included and the numbers of genes within each category; see Table
causes of relatively rare conditions (1–3). However, genomic se-
quencing has myriad potential applications in more general clinical
medicine, including in healthy individuals (3–8). Author contributions: B.D.S. designed research; B.D.S., A.-D.N., and K.A.B. performed re-
Despite the promise of the “age of genomic medicine,” a key search; B.D.S., A.-D.N., K.A.B., and T.G.W. contributed new reagents/analytic tools; B.D.S.,
barrier to translating the power of genomic sequencing to the A.-D.N., K.A.B., and T.G.W. analyzed data; and B.D.S. and T.G.W. wrote the paper.
general clinical setting involves the time and resources required for The authors declare no conflict of interest.
clinically relevant analysis beyond searching for the cause of a single, This article is a PNAS Direct Submission.
usually relatively severe, disease. A number of freely or commer- 1
To whom correspondence should be addressed. E-mail: [email protected].
cially available tools allow curation of individual genomes, including This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
analysis of variant type, predicted pathogenicity of a particular 1073/pnas.1302575110/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1302575110 PNAS | June 11, 2013 | vol. 110 | no. 24 | 9851–9855

Table 1. Organization of the 2,616 genes included in the by Personal Genome Project (www.personalgenomes.org), contain
Manifestation categories and Intervention categories robustly annotated variant sets and admirably enable powerful
Number of genes Number of genes analysis of individual genomes (6). However, this latter type of
in Manifestation in Intervention application focuses on broader genomic exploration, especially
Category categories* categories† involving genetic risk factors and association-based susceptibilities.
This type of analysis tool may be more cumbersome when con-
Allergy/Immunology/Infectious 272 251 sidering the optimal clinical approach to genetic findings in real
Audiologic/Otolaryngologic 273 150 time. Furthermore, the Personal Genome Project has deliberately
Biochemical 452 175 selected an evaluation strategy involving a peer-produced model
Cardiovascular 486 267 that will fill in a “genomic scaffold,” intentionally resulting in the
Craniofacial 384 0 steady and ongoing construction of the analysis platform through
Dental 95 0 the input of multiple contributors (6).
Dermatologic 391 33 We do anticipate that the CGD will serve as a tool that can be
Endocrine 300 153 used in conjunction with some of the above-mentioned plat-
Gastrointestinal 393 106 forms, as well as with other related resources. Admittedly, a key
General 45 1,291 to advancing the field of genomic medicine involves merging the
Genitourinary 183 27 strengths of different resources, including both clinical and ge-
Hematologic 326 247 netic/genomic datasets (14, 15). To expedite and encourage this
Musculoskeletal 801 43 process, the most recently updated contents of the CGD are
Neurologic 1183 46 freely downloadable through the Web site, either in its entirety
Obstetric 38 35 or as selected portions (e.g., including entire clinical categories).
Oncologic 230 229 The full content (current as of April 2013) is also available as
Ophthalmologic 566 51 a single file here (Dataset S1).
Pharmacogenomic 0‡ 186 One way in which the CGD is different from other resources,
Pulmonary 96 60 which may in some respects be advantageous, is that a single
Renal 355 133 (board-certified) clinical geneticist manually investigated and
applied the same rationale to the availability of interventions for
Values shown may differ from updated versions available on the CGD
website. To reflect the multisystemic nature of many genetic disorders and
each gene/condition. Nevertheless, further large-scale validation
to allow comprehensive browsing, each entry may be listed under multiple and testing will be needed, and the CGD must be flexible to in-
categories. corporate new data (16, 17). In general, databases and tools like
*Manifestation categories include organ systems that are primarily affected the CGD must be designed to evolve with the pace of genetic/
by mutations in the corresponding gene. Recognition of these affected sys- genomic and more general medical discovery. Reasons necessi-
tems may aid in condition recognition, as well as supportive care. Genes not tating a dynamic database include, but are not limited to, new
categorized by organ systems within the Manifestation categories are in- discoveries of genetic sources of disease, development of novel
cluded in the General category here.
†
treatment methods, and new clinically relevant findings in pre-
Intervention categories include organ systems for which specific medical
viously described disorders. Without active maintenance, resources
interventions are available. Genes not meeting the described criteria for
these specific interventions (see Materials and Methods) are included in
like the CGD will become almost immediately obsolete.
the General category here. It is important to point out that the lack of available objective
‡
Pharmacogenomic-related genes are all categorized under the Intervention and uniform data related to each gene and condition raises chal-
categories rather than Manifestation categories. lenges. One reason for this subjectivity is that many conditions have
been described in only a small number of individuals, making
drawing large-scale evidence-based conclusions difficult. Never-
2 for an example of a single entry. The entire current contents (as theless, rare conditions are just as important to the individuals they
of April 2013) are available as Dataset S1. The most current ver- affect as common conditions, and in the age of “personalized ge-
sions of selected data (e.g., individual or multiple entries or cate- nomic medicine,” the individual-scale is at the forefront.
gories), as well as the entire contents, can also be freely downloaded In other words, the rarity and nature of many Mendelian
directly through the CGD Web site. conditions make classic “randomized, placebo-based, double-
blind” studies very difficult or impossible. As relates to the cre-
Discussion ation of the CGD, despite attempts to apply uniform criteria in
The clinical interpretation of genomic data involves multiple our analysis, determinations of which conditions have available
complex and controversial issues, and the body of literature on specific interventions are clearly subjective. This subjectivity
the subject is large and rapidly growing. One key challenge stems not only from medical judgment about the condition and
involves the ability to efficiently analyze vast amounts of data in its manifestations, but also from an assessment of the available
a medically meaningful way. This issue will grow as large-scale interventions (16–19). One way to address this problem, as several
sequencing becomes more frequently used in clinical situations. investigations have done (6, 11, 16, 18), is to introduce a semi-
We expect that the creation and ongoing updating of the CGD to quantitative rating system, but adding a numerical score may not
maintain currency will contribute to address this challenge. always provide adequate justification.
We feel that the CGD fills a currently largely unfilled but critical There is also the perhaps larger problem related to the in-
niche in the field of clinical genomics and genomic medicine. terpretation of the potential pathogenicity of specific variants,
Resources, such as Online Inheritance in Man (OMIM) (www. including in genes and conditions where the availability and ben-
omim.org), provide vast repositories of rich clinical and genetic efit of early interventions is less in question (17, 18, 20). Even in
knowledge, but may be harder to query for efficient clinically ori- the situation of relatively well-characterized conditions, such as
ented analysis. Variant-related databases, including the Human phenylketonuria [resulting from mutations in phenylalanine hy-
Gene Mutation Database (HGMD) (www.hgmd.cf.ac.uk and www. droxylase (PAH)] (21–23) or high-penetrance cancer predisposition
biobase-international.com/product/hgmd) are valuable when con- conditions (such as Lynch syndrome) (24–28), a genotype-first
sidering the potential pathogenicity of detected genetic variants, approach may be problematic because of challenges related to
but do not focus directly on clinical implications in a particular predicting the functional and clinical consequences of a detected
healthcare situation. Other databases and tools, such as the variant. In other words, evaluating the potential benefits of

9852 | www.pnas.org/cgi/doi/10.1073/pnas.1302575110 Solomon et al.

Table 2. Sample CGD entry for one gene
Entrez Age Allelic Manifestation Intervention
Gene Gene ID Condition Inheritance group conditions categories categories Interventions/Rationale References

CTNS 1497 Cystinosis AR Pediatric Biochemical; Biochemical; Cystine-depleting agents 6038997;

Endocrine; Endocrine; (cysteamine) instituted at 4914142;
Musculoskeletal; Ophthalmologic; an early age can be beneficial 5443335;
Ophthalmologic; Renal related to manifestations 406375;
Renal affecting multiple organ 333912;
systems, perhaps including 7112129;
cognitive development, as 3307383;
well as renal manifestations 3335962;
(renal tubular Fanconi 3550461;
syndrome and glomerular 3674101;
damage), although renal 3821824;
transplantation may be 3335962;
necessary; Surveillance for 3185663;
renal manifestations and 3292915;
related sequelae can allow early 381441;
detection and management 2230837;
of disease with (in addition to 552398;
cystine-depleting agents), 8455682;
replacement of renal losses; 8172256;
Dietary management (e.g., 7593434;
ensuring sufficient caloric 9537412;
intake, and with vitamin D and 10556299;
phosphate supplementation) 10417278;
can be beneficial related to 10444339;
potential failure to thrive and 10625078;
hypophosphatemic rickets; 11001803;
Cysteamine eyedrops can 10673275;
be beneficial related to 12110740;
ophthalmologic sequelae; 12442267;
Surveillance for endocrine 16603246;
manifestations (e.g., 17643777;
hypothyroidism, or 19863563;
hypogonadism in males) 20301574;
can allow early detection 20803298;
and medical management, 20814825;
including potential use of 21305353;
growth hormone to optimize 21784456;
height in some individuals 21868618;
21371554;
21900880;
22903658;
23001048;
23462307;
23538568

GENETICS
The CGD can be queried with a single gene symbol or condition, or with a list. Searches can also be limited by organ-system–based clinical categories (for
both Manifestation and Intervention categories). To allow efficient access to complementary resources, including databases of variants and apparently
pathogenic mutations, each gene/condition-specific links to the relevant entry (where available) in: 1000 Genomes (www.1000genomes.org); the Short
Genetic Variations Database (The Database of Single Nucleotide Polymorphisms, dbSNP) (www.ncbi.nlm.nih.gov/snp/); GeneTests (www.ncbi.nlm.nih.gov/
sites/GeneTests/); the HGMD (www.hgmd.cf.ac.uk/); the National Center for Biotechnology Information gene database (www.ncbi.nlm.nih.gov/gene); the
National Heart, Lung, and Blood Institute Gene Ontology Exome Sequencing Project Exome Variant Server (http://evs.gs.washington.edu/EVS); OMIM (http://
www.omim.org); Genetics Home Reference (http://ghr.nlm.nih.gov/); ClinVar (http://www.ncbi.nlm.nih.gov/clinvar/). Each reference is directly linked to
a PubMed abstract (>20,000 articles are individually referenced).
Information shown above may differ from the updated version available on the CGD website. For the sake of viewability, the “Comments” column (which
appears on the CGD website) has been omitted from this example entry.

genetic/genomic diagnosis (i.e., before diagnosis on clinical a truncating mutation or a large deletion in a gene in which loss-
grounds alone) of a disorder is challenging in itself, but is only one of-function is predicted to cause disease. The second reason, used
piece of the puzzle. An equally large piece involves the evaluation for variants whose pathogenicity is more difficult to assess (e.g.,
of specific variants, and strategies have been devised to address novel missense variants), would involve the presence of strong
this issue. One such strategy involves “binning” variants (11, 18). evidence showing that the exact detected variant had been pre-
According to this type of algorithm, a variant would be judged to viously established as disease-causing (11, 12, 18, 20).
be pathogenic for one of two reasons. The first reason is that the Along these lines, in any approach involving the clinical man-
type of variant may by itself predict pathogenicity: for example, agement of individual patients, genotypic information should not

Solomon et al. PNAS | June 11, 2013 | vol. 110 | no. 24 | 9853
operate in a vacuum (6, 29). Just as genotypic data, such as the point, precise diagnosis is challenging for many conditions, and correct
type, location, and novelty of a particular amino acid substitution recognition based on genetic/genomic diagnosis may allow interventions
can aid in the interpretation of a variant, clinical information, related to specific manifestations. The efficacy of these interventions would
be diminished or lost with later diagnosis, such as might occur based pri-
including family and medical history, can help determine the
marily upon clinical presentation. For example, in certain types of Ehlers-
consideration of a specific variant (5, 12, 29). Danlos syndrome, which may not always be recognized early enough to
The first goal of the CGD is broad dissemination to solicit allow optimal medical care, genotype-based recognition may allow inter-
content-oriented feedback and input from experts studying rele- ventions related to certain cardiovascular manifestations, which may reduce
vant genes and conditions. This input can be used to continually associated morbidity and mortality (31).
revise and improve this resource, as the CGD will be regularly For the Intervention categories, all genes not meeting the above criteria
updated through a combination of automated and manual cura- were included in the General category. As described above, for many such
tion. The first long-term objective is the establishment of a user- conditions although a more specific intervention may not be currently
friendly resource relevant to a wide group of clinicians that can be available, genetic knowledge may be beneficial related to a number of issues,
used as a reference resource in a variety of situations. Eventually, including the selection of optimal supportive care, prognostic considerations
in addition to serving as a reference tool, the CGD may be used related to medical-decision making, informing reproductive decisions, and
avoidance of unnecessary testing as part of the diagnostic process. These
as a filter superimposed on automated binning algorithms to help
entries contain similar information to those classified by organ system, with
allow efficient, clinically relevant annotation of human genomes. the exception that the interventions and rationale are not specifically de-
scribed. Individual experts were contacted in many instances where the
Materials and Methods availability or efficacy of interventions was unclear.
To investigate conditions with identified genetic underpinnings, we indi- The Web interface to the CGD allows searching by gene or condition or
vidually read all entries in OMIM (www.omim.org) that included conditions browsing by categories. For each entry, the database includes the gene
with genetic causes, then cross-referenced all entries—and searched for ad- symbol, conditions, allelic conditions (conditions resulting from mutations
ditional entries—within the following publicly available databases: GeneTests in the same gene, but which themselves may not have a specific intervention
(www.ncbi.nlm.nih.gov/sites/GeneTests), Pharmacogenomics Knowledge Base
available; it must be noted that for many reportedly distinct conditions, there
(www.pharmgkb.org), HGMD (www.hgmd.cf.ac.uk), and HGMD Professional
is clearly a phenotypic continuum, such that division into clinically separate
(https://portal.biobase-international.com/hgmd) through a site-specific li-
conditions can be challenging), clinical categorization (as described above, by
cense. Pertaining to each gene and condition described in these databases,
both manifestations as well as more specific interventions), inheritance, age
we directly analyzed the content of all cited primary references. Published
[designated as either pediatric (less than 18 y of age) or adult] in which
literature that was not included in these databases was also queried through
interventions are indicated based on descriptions in the medical literature,
independent PubMed search (by gene and condition name). The most recent
and general descriptions of the interventions/rationale. This latter category is
date of query was April 17, 2013.
not intended to serve in place of comprehensive treatment guidelines nor act
The CGD has been constructed to reflect the multisystemic nature of many
as a clinical guide, but rather briefly describes the types of interventions that
genetic conditions to allow more comprehensive browsing by clinical cate-
may be considered.
gories. In the CGD, genes were first categorized into Manifestation categories,
The CGD currently includes only single gene alterations; it does not include
or the organ systems primarily affected by mutations in the corresponding
contiguous gene syndromes, although conditions with, for example, dem-
gene. For many of these organ systems, recognition of the condition’s effects
onstrated digenic inheritance are included. Similarly, somatic alterations,
and related supportive care may be clinically beneficial. Conditions not
grouped within a specific organ system under the Manifestation categories such as commonly occur in cancerous processes, are not included, although if
were included in the General category. a germ-line change in the same gene has been shown to result in disease,
Next, genes were separately categorized under Intervention categories by those latter conditions are included. The current version does not include
the organ systems for which specific medical interventions were available. In susceptibilities or genetic associations, such as those identified through an
determining the Intervention categories, the following points were consid- association-based study. As the database expands in the future, these types
ered. These points are based in part on arguments related to the selection of of additions would be considered.
targets for routine newborn screening (30): (i) the condition must be clini-
cally significant (i.e., at least some manifestations must result in morbidity or ACKNOWLEDGMENTS. The authors thank Leslie G. Biesecker, Derek A. T.
mortality); (ii) there must be a currently available, potentially beneficial Cummings, James P. Evans, Donald W. Hadley, and Maximilian Muenke for
support, mentorship, and critical input; Andreas D. Baxevanis for bioinfor-
intervention (this intervention may include preventive measures, surveil-
matic discussions and support; Mark Fredriksen for programming assistance;
lance, or medical or surgical treatments, although experimental/research- and all the experts who provided input related to individual genes and
based interventions were not included); (iii) there should be advantage to conditions. This research was supported by the Intramural Research Program
early (genomic) diagnosis as opposed to discovery of the condition on purely of the National Human Genome Research Institute, National Institutes
clinical grounds (i.e., without genetic/genomic testing). Regarding this last of Health.

1. Bamshad MJ, et al. (2011) Exome sequencing as a tool for Mendelian disease gene 11. Berg JS, Khoury MJ, Evans JP (2011) Deploying whole genome sequencing in clinical
discovery. Nat Rev Genet 12(11):745–755. practice and public health: Meeting the challenge one bin at a time. Genet Med 13(6):
2. Biesecker LG, Shianna KV, Mullikin JC (2011) Exome sequencing: The expert view. 499–504.
Genome Biol 12(9):128. 12. Solomon BD, et al.; NISC comparative Sequencing Program (2012) Incidental medical
3. Gonzaga-Jauregui C, Lupski JR, Gibbs RA (2012) Human genome sequencing in health information in whole-exome sequencing. Pediatrics 129(6):e1605–e1611.
and disease. Annu Rev Med 63:35–61. 13. Teer JK, Green ED, Mullikin JC, Biesecker LG (2012) VarSifter: Visualizing and ana-
4. Bainbridge MN, et al. (2011) Whole-genome sequencing for optimized patient lyzing exome-scale sequence variation data on a desktop computer. Bioinformatics
management. Sci Transl Med 3(87):87re3. 28(4):599–600.
5. Solomon BD, et al.; NISC Comparative Sequencing Program (2011) Personal- 14. Oetting WS, et al. (2013) Getting ready for the Human Phenome Project: The 2012
ized genomic medicine: Lessons from the exome. Mol Genet Metab 104(1-2): Forum of the Human Variome Project. Hum Mutat 34(4):661–666.
189–191. 15. Hamosh A, et al. (2013) PhenoDB: A new web-based tool for the collection, storage,
6. Ball MP, et al. (2012) A public resource facilitating clinical use of genomes. Proc Natl and analysis of phenotypic features. Hum Mutat 34(4):566–571.
Acad Sci USA 109(30):11920–11927. 16. Green RC, et al. (2012) Exploring concordance and discordance for return of incidental
7. Johnston JJ, et al. (2012) Secondary variants in individuals undergoing exome se- findings from clinical sequencing. Genet Med 14(4):405–410.
quencing: Screening of 572 individuals identifies high-penetrance mutations in can- 17. Manolio TA, et al. (2013) Implementing genomic medicine in the clinic: The future is
cer-susceptibility genes. Am J Hum Genet 91(1):97–108. here. Genet Med 15(4):258–267.
8. Solomon BD, Pineda-Alvarez DE, Bear KA, Mullikin JC, Evans JP; NISC Comparative 18. Berg JS, et al. (2013) An informatics approach to analyzing the incidentalome. Genet
Sequencing Program (2012) Applying genomic analysis to newborn screening. Mol Med 15(1):36–44.
Syndromol 3(2):59–67. 19. Evans JP, Berg JS (2011) Next-generation DNA sequencing, regulation, and the limits
9. Lupski JR, et al. (2010) Whole-genome sequencing in a patient with Charcot-Marie- of paternalism: The next challenge. JAMA 306(21):2376–2377.
Tooth neuropathy. N Engl J Med 362(13):1181–1191. 20. Xue Y, et al.; 1000 Genomes Project Consortium (2012) Deleterious- and disease-allele
10. Tong P, et al. (2010) Sequencing and analysis of an Irish human genome. Genome Biol prevalence in healthy individuals: Insights from current predictions, mutation data-
11(9):R91. bases, and population-scale resequencing. Am J Hum Genet 91(6):1022–1032.

9854 | www.pnas.org/cgi/doi/10.1073/pnas.1302575110 Solomon et al.

21. DiLella AG, Kwok SC, Ledley FD, Marvit J, Woo SL (1986) Molecular structure and 26. Papadopoulos N, et al. (1994) Mutation of a mutL homolog in hereditary colon
polymorphic map of the human phenylalanine hydroxylase gene. Biochemistry 25(4): cancer. Science 263(5153):1625–1629.
743–749. 27. Miyaki M, et al. (1997) Germline mutation of MSH6 as the cause of hereditary non-
22. Holtzman NA, Kronmal RA, van Doorninck W, Azen C, Koch R (1986) Effect of age at polyposis colorectal cancer. Nat Genet 17(3):271–272.
28. Hampel H, et al. (2005) Screening for the Lynch syndrome (hereditary nonpolyposis
loss of dietary control on intellectual performance and behavior of children with
colorectal cancer). N Engl J Med 352(18):1851–1860.
phenylketonuria. N Engl J Med 314(10):593–598.
29. Spencer DH, et al. (2011) Direct-to-consumer genetic testing: Reliable or risky? Clin
23. Drogari E, Smith I, Beasley M, Lloyd JK (1987) Timing of strict diet in relation to fetal
Chem 57(12):1641–1644.
damage in maternal phenylketonuria. An international collaborative study by the
30. American College of Medical Genetics’ Newborn Screening Expert Group (2006)
MRC/DHSS Phenylketonuria Register. Lancet 2(8565):927–930. Newborn screening: Toward a uniform screening panel and system. Genet Med
24. Fishel R, et al. (1993) The human mutator gene homolog MSH2 and its association 8(Suppl 1):1S–252S.
with hereditary nonpolyposis colon cancer. Cell 75(5):1027–1038. 31. Schwarze U, et al. (2004) Rare autosomal recessive cardiac valvular form of Ehlers-
25. Nicolaides NC, et al. (1994) Mutations of two PMS homologues in hereditary non- Danlos syndrome results from mutations in the COL1A2 gene that activate the
polyposis colon cancer. Nature 371(6492):75–80. nonsense-mediated RNA decay pathway. Am J Hum Genet 74(5):917–930.

GENETICS

Solomon et al. PNAS | June 11, 2013 | vol. 110 | no. 24 | 9855