Review paper
Immunodeficiency information services

Introduction
Primary immunodeficiency diseases (IDs) consist of a group of inherited, often rare disorders affecting the immune system. They predispose individuals to various clinical symptoms such as recurrent and persistent infections, allergies, cancer and autoimmune manifestations [1, 2]. An immune defect can affect any part of the immune system. This has been used as the basis of the ID classification (fig. 1). A wealth of information concerning these rare diseases is available from the Internet, but it is scattered and very fragmented. We have compiled all the major immunodeficiency data and thousands of links onto one server, the Immunodeficiency Resource (IDR) [3, 4]. Fact files for each disease serve as the core of the IDR [5]. The server offers an extensive starting point for immunodeficiency information retrieval. Around 150 primary immunodeficiency diseases have been identified since the 1950’s. The molecular basis of more than 100 primary IDs have recently been identified [6]. Inheritance of primary IDs can be X-linked, autosomal recessive, or autosomal dominant (table 1). Diagnosis of an immunodeficiency can be difficult because similar symptoms characterise several disorders and mutations from the same gene can lead to distinct phenotypic consequences [7, 8]. Mutation detection is the most reliable method to confirm the diagnosis. New immunodeficiency-related genes and new mutations and patients are frequently found. We have established immunodeficiency mutation databases (IDbases) for most of the IDs for which the affected gene has been identified [2, 9] (table 2). These databases enable one to carry out detailed mutation studies and handle the ever-growing volume of information. For the treatment and survival of the immunodeficiency patient, it is important to have the correct diagnosis as soon as possible. Because many IDs are very rare disorders there are only limited number of laboratories carrying out genetic testing for patients and it may be difficult for a physician to find the right laboratory. We maintain an online registry of genetic and clinical immunodeficiency diagnostic laboratories [10]. The IDdiagnostics service is primarily directed for health care professionals enabling them to find a laboratory to contact for the required genetic and/or clinical diagnostics. The aim of the registry is also to increase the general awareness of IDs, which is important for fast and reliable diagnosis and proper treatment. Diagnosis of the IDs with overlapping symptoms is often troublesome. PIDexpert is our most recent instrument for faster and more accurate diagnosis. It is a medical expert system designed to give the diagnostic picture of IDs based on symptoms, signs, medical history, physical findings and laboratory tests. The system is expected to contribute significantly to the diagnosis and treatment of patients and save money in health care. The PIDexpert system will be available on our server in the near future.
Immunodeficiency resource (IDR)
The IDR is a compendium of information on immunodeficiencies. It is freely available online via the Internet (http://bioinf.uta.fi/idr/) [3, 4]. It is maintained for collecting and distributing all the essential information and links related to immunodeficiencies. The Internet encompasses billions of web pages including a lot that contain questionable, even misleading data, and it may be hard to distinguish the relevant data from nonsense. The validation of information in the IDR is of prime importance. Experts check the data and approve only the data and sites with solid scientific and medical information to be linked to the IDR [4]. Links to external information sources are checked especially carefully by curators before being accepted. The contents of the IDR are versatile. The IDR offers tens of thousands of validated links to other sites, which are periodically automatically checked. Navigation is logical and it is possible to search for any text string or multiple strings with Boolean logic across the IDR pages. The IDR is designed for different user groups: researchers, physicians, and nurses, as well as patients and their families and the general public. The general introduction pages are extensive. From the IDR one can also find comprehensive classification and diagnostic criteria for some immunodeficiencies [8]. Genes causing immunodeficiencies are listed in accordance to disease classification, with hyperlinks to several bioinformatics databases. Further, there is information about analysis, reference sequences, protein structures, animal models and knock-outs, and a picture gallery. There are several societies related to immunodeficiency research, treatment and care and patients. We have collected links to immunology societies, nurse societies and patient organizations. We also maintain a list of current immunology meetings and workshops. The immunology laboratories page contains a list of home pages of laboratories that are active in many fields of immunodeficiency research. Patient pages have useful links for immunodeficiency patients and their families, containing mostly personal reports about surviving and living with an immunodeficiency and contact information with some ID communities. The online library and Immunology resources provide further information about immunology and IDs. The IDR pages are extensively hyperlinked to our on-line immunology glossary of over 1,000 immunological terms. Descriptions for abbreviations and acronyms used in IDR pages are also available. The basic information on IDs is stored as fact files [4]. Each fact file provides a succinct summary of information on one disease and the affected gene, including clinical information, molecular biology, and HTML hyperlinks to other Internet resources. Fact files are a compact package of information, and a good starting point for expansive data searching related to IDs. Currently we have 135 fact files. The fact files are created by using the IDML data model [4, 5]. IDML (Inherited Disease Markup Language) is based on the eXtensible Markup Language (XML) format. IDML integrates biomedical information related to hereditary diseases into a Web and WAP accessible knowledge base. The data model has been applied to primary immunodeficiencies, but it can be used for any group of hereditary diseases. The fact files are also accessible with mobile devices by using the BioWAP (http://bioinf.uta.fi/BioWAP/) service [11, 12].
Immunodeficiency mutation databases (IDbases)
Immunodeficiency mutation databases (IDbases) offer an easy way to find recently identified mutations, to compare genotype-phenotype correlations, and to discover a specific mutation or to examine the most common mutations in a single immunodeficiency related gene. We currently maintain databases for 110 IDs with more than 4100 public patient entries at IMT Bioinformatics (http://bioinf.uta.fi/IDbases/) (table 1). Databases are named according to the affected gene after systematic HUGO Gene Nomenclature (http://www.gene.ucl.ac.uk/nomenclature/). The IDbases are patient-related databases, where the mutation data has been collected into entries along with some clinical information. This allows the discovery of statistically relevant trends from large data sets. Patient-related mutation data makes it possible to find out which mutations are frequent among patients with certain symptoms. In addition to the actual mutations and clinical information, every database has further information about mutation types and visualisation of the distribution of mutations within the amino acid sequence. The pages are interactive, providing access to patient information, mutation information, and literature references. It is easy to trace all the publications related to a specific mutation. There are also links to reference sequences and other data sources, such as sequence databanks, GeneCard, OMIM and UniGene. In addition to our databases we have links to immunodeficiency databases maintained by others. Mutation data submissions from the scientists analysing mutations are advantageous for keeping the databases up-to-date. Mutation data may be submitted either by contacting the curators or via the Internet. Each database has a specific electronic submission form. The interactive submission form, based on the MUTbase program [13], facilitates submission of the mutation. The program compares the given mutation information to standard reference sequences provided by the IDRefSeq (http://bioinf.uta.fi/IDRefSeq/) service, and warns of possible errors. It calculates the protein level change(s) caused by the mutation(s) and checks for the numbering and type of the nucleotide(s) affected. The MUTbase system generates a standardized representation of the information contained in the raw data, part of which is added to the database entry and the other part written on numerous interactive Web pages. Finally, the submission is sent to the curators by e-mail. The curator of the database validates all the submitted data before it is made public. The European Society for Immunodeficiencies (ESID) registry (http://www.esid.org/esid_registry.php) has collected information about immunodeficiency patients. The database has undergone a complete rebuild during last two years. BTKbase for X-linked agammaglobulinemia (XLA) was the first IDbase established in 1994 [14-16]. Today the cooperation with the patient database continues, productively integrating the IDbase mutation service with the ESID registry. The ESID registry collects new patient information containing confidential clinical data and at the same time receives some new mutations. The mutations are validated by the IDbase submission system. The collaboration facilitates direct submission to both the ESID registry and IDbases, and thus allows both systems to be updated by a single submission.
Immunodeficiency diagnostics registry (IDdiagnostics)
Early and reliable diagnosis is often crucial for the efficient treatment of IDs. If diagnosis and treatment are delayed, it may even cost a patient’s life. For most IDs, detecting the molecular defect is essential for the correct diagnosis. The number of the laboratories analysing the genetic defects of IDs is limited, due to rareness of immunodeficiencies. IDdiagnostics is a registry of laboratories performing genetic testing for patients with hereditary immunodeficiencies [3]. It is formed from two independent registries for laboratories performing genetic and clinical tests for IDs, respectively [10]. These registries provide a service for those trying to find the nearest and/or most suitable laboratory conducting ID testing. IDdiagnostics currently contains information for the analysis of defects in 41 ID-related genes. The service is intended for physicians, researchers, and other health professionals involved with medical genetics. Laboratories are included in IDdiagnostics on a voluntary basis. The registry includes only those laboratories willing to have their information posted on the Internet. Registration forms are available in both electronic and paper form. The IDdiagnostics data is regularly updated and laboratories are contacted to verify the accuracy of their information. The curators retain the right to remove information for a laboratory if there are problems, e.g., with the time schedule or quality of information. Physicians have to contact the laboratory before sending in any samples as the standards vary between laboratories. The cost of the analysis varies depending, for example, on the method used, the type of laboratory, and the research interest of a particular disease. The gene test laboratories provide information about the time required for a diagnosis and the turnaround time, how often the samples are run, how many samples are studied annually and the cost of the analyses. Contact addresses for laboratories performing diagnosis are provided in the IDdiagnostics registry along with the assay method(s) used. The search facilities of the IDdiagnostics database allow users to run text based search queries. Gene test laboratories can be searched by disease name (including alternative names), gene symbol, OMIM code, laboratory name, laboratory location and free text. A search engine facilitates finding laboratories for certain disease(s), methods and/or geographical locations. Further information and submission pages for both genetic and clinical testing can be found at http://bioinf.uta.fi/IDdiagnostics/.
PIDexpert
Medical expert systems (MESs) or medical diagnostic decision support systems (MDDS) [17] are an established component of medical technology. They are computer programs that use a set of rules applied to knowledge extracted from human experts. Medical expert systems help in diagnostic processes and report generation, improve consistency in decisions, and increase timeliness in decision-making and productivity. In medicine, expert systems have been used in a variety of fields, such as internal medicine, paediatrics, infectious disease, neurology, psychiatry. Medical expert systems vary in complexity. They produce patient-specific and situation specific recommendations. MESs can be integrated with other applications, such as electronic patient records, systems for prescribing and dispensing medicines, and other information systems used in health settings. PIDexpert is a medical expert system, which aims to help with the diagnosis and management of primary immunodeficiency diseases. It can act as both an electronic textbook and an expert consultant program. PIDexpert generates a differential diagnosis from clinical symptoms, provides justification for a diagnosis and suggests potentially useful further clinical information that is required. PIDexpert includes a knowledge acquisition system, a knowledge base, an inference engine and a user interface [Samarghitean and Vihinen, submitted]. The knowledge base is built with data and facts from IDR, IDdiagnostics, literature and medical experts. Additionally, real examples of differential diagnoses of patient cases will further enrich the knowledge base. The ESID/PAGID diagnostic guidelines [8] and practice parameters for the diagnosis and management of primary immunodeficiencies established by the American College of Allergy, Asthma and Immunology (ACAAI) and the Joint Council of Allergy, Asthma and Immunology [18] will also be included in the knowledge base. These guidelines provide heuristics for possible/probable/definitive diagnosis for some of the most common IDs. The inference engine includes the rules or facts used for deduction and in this case use decision tree algorithm. This algorithm is well suited due to its symbolic knowledge representation and explanations of decisions it makes. Decision tree methodology does not require as many known cases as some other artificial intelligence (AI) methods, and that is convenient in the case of rare IDs. The user-friendly interface will be web-based. The patient information is not submitted over the Internet because Java technology allows the analysis to run on the local computer. Physicians use an input form to indicate various signs and symptoms. The system also identifies other conditions that are associated with the disorder and how the diagnosis can be confirmed. If necessary, the program will also remind or suggest an additional test. The AI system is not intended to replace the physician, but to help in decision making. PIDexpert is still under development, but it will be available in the near future (http://bioinf.uta.fi/PIDexpert/).
Conclusions
The Internet provides an effective way to search information on countless databases and web pages, even for rare diseases like primary IDs. All the data found on the Internet is not qualified, and there is clearly a need for a reliable and comprehensive knowledge base for primary IDs. The content of our services has been extended over the recent years. The frequently expanding and detailed data of the IDR, the IDbases, and IDdiagnostics increasingly serve the needs of diverse user groups, such as physicians, nurses, researchers and patients as well as their families in their fight against primary IDs. The services also increase the common knowledge of primary IDs, which encourages physicians to consider the possibility of a primary ID, when warranted. This is important, because an early diagnosis provides better prognosis for the patient. The PIDexpert system will complement the services by providing a useful and efficient instrument for problematic ID diagnosis.
Acknowledgements
The financial support from the EU, the Finnish Academy, and the Medical Research Fund of Tampere University Hospital is gratefully acknowledged.
References
1. Fischer A (2004): Human primary immunodeficiency diseases: a perspective. Nat Immunol; 5: 23-30. 2. Vihinen M, Arredondo-Vega FX, Casanova JL, et al (2001): Primary immunodeficiency mutation databases. Adv Genet; 43: 103-188. 3. Väliaho J, Riikonen P, Vihinen M (2000): Novel immunodeficiency data servers. Immunol Rev; 178: 177-185. 4. Väliaho J, Pusa M, Ylinen T, Vihinen M (2002): IDR: the ImmunoDeficiency Resource. Nucleic Acids Res 30: 232-234. 5. Väliaho J, Riikonen P, Vihinen M (2005): Distribution of immunodeficiency fact files with XML – from Web to WAP. BMC Med Inform Decis Mak; 5: 21. 6. Notarangelo L, Casanova JL, Fischer A, et al (2004): Primary immunodeficiency diseases: an update. J Allergy Clin Immunol 114: 677-687. 7. Fischer A (2001): Primary immunodeficiency diseases: an experimental model for molecular medicine. Lancet 357: 1863-1869. 8. Conley ME, Notarangelo LD, Etzioni A (1999): Diagnostic criteria for primary immunodeficiencies. Representing PAGID (Pan-American Group for Immunodeficiency) and ESID (European Society for Immunodeficiencies). Clin Immunol 93: 190-197. 9. Lappalainen I, Ollila J, Smith CIE, Vihinen M (1997): Registries of immunodeficiency patients and mutations. Hum Mutat 10: 261-267. 10. Samarghitean C, Väliaho J, Vihinen M (2004): Online registry of genetic and clinical immunodeficiency diagnostic laboratories, IDdiagnostics. J Clin Immunol 24: 53-61. 11. Riikonen P, Boberg J, Salakoski T, Vihinen M (2001): BioWAP, mobile Internet service for bioinformatics. Bioinformatics 17: 855-856. 12. Riikonen P, Boberg J, Salakoski T, Vihinen M (2002): Mobile access to biological databases on the Internet. IEEE Trans Biomed Eng 49: 1477-1479. 13. Riikonen P, Vihinen M (1999): MUTbase: maintenance and analysis of distributed mutation databases. Bioinformatics 15: 852-859. 14. Vihinen M, Cooper MD, de Saint Basile G, et al (1995): BTKbase: a database of XLA-causing mutations. International Study Group. Immunol Today 16: 460-465. 15. Vihinen M, Iwata T, Kinnon C, et al (1996): BTKbase, mutation database for X-linked agammaglobulinemia (XLA). Nucleic Acids Res 24: 160-165. 16. Vihinen M, Kwan SP, Lester T, et al (1999): Mutations of the human BTK gene coding for bruton tyrosine kinase in X-linked agammaglobulinemia. Hum Mutat 13: 280-285. 17. Miller RA (1994): Medical diagnostic decision support systems-past, present, and future: a threaded bibliography and brief commentary. J Am Med Inform Assoc 1: 8-27. 18. Bonilla FA, Bernstein IL, Khan DA, et al (2005): Practice parameter for the diagnosis and management of primary immunodeficiency. Ann Allergy Asthma Immunol 94: S1-63. 19. Concannon P, Gatti RA (1997): Diversity of ATM gene mutations detected in patients with ataxia-telangiectasia. Hum Mutat 10: 100-107. 20. Rong SB, Väliaho J, Vihinen M (2000): Structural basis of Bloom syndrome (BS) causing mutations in the BLM helicase domain. Mol Med 6: 155-164. 21. Touitou I, Lesage S, McDermott M, et al (2004): Infevers: an evolving mutation database for auto-inflammatory syndromes. Hum Mutat 24: 194-198. 22. Notarangelo LD, Peitsch MC (1996): CD40lbase: a database of CD40L gene mutations causing X-linked hyper-IgM syndrome. Immunol Today 17: 511-516. 23. Saunders RE, Goodship TH, Zipfel PF, Perkins SJ (2006): An interactive web database of factor H-associated hemolytic uremic syndrome mutations: insights into the structural consequences of disease-associated mutations. Hum Mutat 27: 21-30. 24. Roos D (1996): X-CGDbase: a database of X-CGD-causing mutations. Immunol Today 17: 517-521. 25. Heyworth PG, Curnutte JT, Rae J, et al (2001): Hematologically important mutations: X-linked chronic granulomatous disease (second update). Blood Cells Mol Dis 27: 16-26. 26. Lappalainen I, Vihinen M (2002): Structural basis of ICF-causing mutations in the methyltransferase domain of DNMT3B. Protein Eng 15: 1005-1014. 27. Levran O, Erlich T, Magdalena N, et al (1997): Sequence variation in the Fanconi anemia gene FAA. Proc Natl Acad Sci U S A 94: 13051-13056. 28. Gillio AP, Verlander PC, Batish SD, et al (1997): Phenotypic consequences of mutations in the Fanconi anemia FAC gene: an International Fanconi Anemia Registry study. Blood 90: 105-110. 29. Auerbach AD, Greenbaum J, Pujara K, et al (2003): Spectrum of sequence variation in the FANCG gene: an International Fanconi Anemia Registry (IFAR) study. Hum Mutat 21: 158-168. 30. Kalmar L, Hegedus T, Farkas H, et al (2005): HAEdb: a novel interactive, locus-specific mutation database for the C1 inhibitor gene. Hum Mutat 25: 1-5. 31. Puck JM (1996): IL2RGbase: a database of gamma c-chain defects causing human X-SCID. Immunol Today 17: 507-511. 32. Vihinen M, Villa A, Mella P, et al (2000): Molecular modeling of the Jak3 kinase domains and structural basis for severe combined immunodeficiency. Clin Immunol 96: 108-118. 33. Notarangelo LD, Mella P, Jones A, et al (2001): Mutations in severe combined immune deficiency (SCID) due to JAK3 deficiency. Hum Mutat 18: 255-263. 34. Villa A, Sobacchi C, Notarangelo LD, et al (2001): V(D)J recombination defects in lymphocytes due to RAG mutations: severe immunodeficiency with a spectrum of clinical presentations. Blood 97: 81-88. 35. Lappalainen I, Giliani S, Franceschini R, et al (2000): Structural basis for SH2D1A mutations in X-linked lymphoproliferative disease. Biochem Biophys Res Commun 269: 124-130.