abstracts - rd-connect.eu · swertz ma1, monaco l2 1university medical center groningen, university...

Abstracts

RD-Connect Annual Meeting

16th – 18th April 2018 Athens, Greece

Abstracts Annual Meeting | 16-18 April 2018 | Athens (Greece)

1

Poster Abstracts | in Alphabetical Order

Poster No.

Presenting Author Title Page

Last Name First Name

P28 Badowska Dorota RD-Connect - an FP7 Success Story 30

P14 Blavier André ALFA v3: ALamut Functional Annotations and Predictions

16

P8 Bosio

Capella Gutierrez

Mattia

Salvador

Disease modifiers in Congenital Myasthenic Syndrome: whole genome and transcriptome integration use case

10

P9 Bosio

Capella Gutierrez

Mattia

Salvador

RD-Connect Multiomics Working Group: Pilot work integrating WGS and RNASeq pipelines

11

P15 Carta Claudio Training activities to promote, create and support international and FAIR rare disease registries

17

P7 Casareto Lorena Playing cards with biobanks: a tool for promoting public engagement

9

P29 Dawson John RD-Connect: data sharing and analysis for rare disease research within the integrated platform and through GA4GH Beacon and Matchmaker Exchange

31

P20 Dorboz Imen Mutation in POLR3K cause hypomyelinating leukodystrophy and abnormal ribosomal RNA regulation

22

P13 Fernández José M Modularizing RD-Connect Central Authentication Service components into software containers

15

P27 Hedley Victoria Recommended Practices for Data Standardisation in the Context of the operation of European Reference Networks

29

P12 Jacobsen Annika FAIRification of rare disease patient registries 14

P16 Kodra Yllka Recommendations for improving the quality of rare disease registries

18

P18 Laurie Steve Challenges and successes of sharing data for resolution of rare disease cases: The example of the BBMRI-LPC WES call, in collaboration with EuroBioBank and RD-Connect

20

P2 Leary Becca European Reference Network for rare Neuromuscular Diseases: EURO-NMD, progress so far

4

P1 López Estrella 2016 BBMRI-LPC WES Call: An opportunity to diagnose six unsolved cases within SpainUDP

3

P21 Macek Milan RD-Connect engagement in Central Eastern and Central Europe

23

P24 Merino-Martinez Roxana F2Share: Federation Framework for sharing research data

26

2 Abstracts Annual Meeting | 16-18 April 2018 | Athens (Greece)

P10 Mina Eleni A comparative study on blood and brain HD signatures: Similarities between mouse and human HD gene expression data

12

P17 Papakonstantinou Anastasios Detecting consanguinity based on whole-exome sequencing: integration within the RD-connect genome-phenome analysis platform and impact on the molecular diagnosis of rare disease patients

19

P22 Patrinos George electronic Pharmacogenomics Assistant (ePGA) 24

P3 Pavlovic Sonja Application of NGS Technology and establishment of biobanks improved research and diagnostics of rare diseases - Serbian 5 years experience

5

P23 Salgado David Aix-Marseille University RD-Connect contribution

25

P6 Spalding Dylan The European Genome-phenome Archive: Archiving RD-Connect data

8

P19 Torreri Paola The RD-Connect Registry & Biobank Finder: an online directory for sharing and integrating RD data and samples from RD registries and biobanks

21

P25 Turner Cathy Collaboration in NeurOmics: enabling effective data-sharing and maximising impact in neuromuscular disease

27

P5 van der Tuin Karin The genetic background of non-medullary paediatric thyroid carcinoma (GeNoThyPe)

7

P4 van Enckevort David RD-Connect Sample Catalogue 6

P11 Wang Mary Developing the international EuroBioBank network of rare diseases biobanks within the RD-Connect platform

13


3

P1 2016 BBMRI-LPC WES Call: An opportunity to diagnose six unsolved cases within SpainUDP

López E1, Bermejo E1, Martínez B2, Alonso FJ2, Posada M1

1Institute of Rare Diseases Research, IIER-ISCIII & Centre for Biomedical Network Research on Rare Diseases (CIBERER), Madrid, Spain, 2Institute of Rare Diseases Research, IIER-ISCIII (Madrid, Spain)

Presenting author: Estrella López Contact: [email protected]

The 2016 BBMRI-LPC Whole Exome Sequencing Call offered a great opportunity to genetically diagnose RD patients with DNA from the EuroBioBank. The Institute of Rare Diseases Research (IIER-ISCIII) led one of the 17 awarded projects, entitled “Undiagnosed cases with complex phenotypes including intellectual disability”. Sequencing for this project was conducted at the Centro Nacional de Análisis Genómico (CNAG-CRG, Spain) and was completed by early 2017. On the other hand, phenotypic terms were extracted from clinical documents stored in the Spanish Rare Diseases Registry, mapped to HPO (Human Phenotype Ontology) terms and uploaded into PhenoTips, a software tool useful for collecting standardized phenotypic information. Data was processed through the RD-Connect validated analysis pipeline and was made available to authorized users through its platform once all the required commitments were fulfilled.

Six undiagnosed pediatric cases from SpainUDP were selected for participation in this Call. All of them had complex phenotypes with intellectual disability as a common feature. After an exhaustive study of patients data in the RD-Connect platform by two independent researchers, a consensus was reached to select the candidate variants. These were confirmed by the Sanger technique in all family members.

As a conclusion, thanks to this Call five patients have been diagnosed with the following disorders: Kabuki syndrome 1, SHORT syndrome, Mental retardation X-linked 102, Combined oxidative phosphorylation deficiency 13 and Mental retardation autosomal dominant 32. On the other hand, one patient is still pending on results, although it is expected to reach a diagnosis at the beginning of 2018.


P2 European Reference Network for rare Neuromuscular Diseases: EURO-NMD, progress so far

Evangelista T, Leary R, Hails M, Bushby K on behalf of ERN, EURO-NMD

John Walton Muscular Dystrophy Research Centre, Institute of Genetic Medicine, Newcastle University, UK

Presenting author: Rebecca Leary Contact: [email protected]

EURO-NMD is a European Reference Network for rare neuromuscular diseases (NMDs), a broad group of related disorders that represent a major cause of mortality and lifelong disability in children and adults. NMDs are caused by acquired or genetic defects of motorneurons, peripheral nerves, neuromuscular junctions or skeletal muscle, resulting in muscle weakness and wasting, swallowing and breathing difficulties, and cardiac failure. NMDs are difficult to recognize, with long delays in diagnosis. No curative treatments yet exist for any NMD and their rarity and diversity pose specific challenges for healthcare, research, and for the development and marketing of therapies.

NMDs collectively affect an estimated 500,000 EU citizens and result in significant costs for families and the healthcare system. EURO-NMD unites 61 of Europe’s leading NMD clinical and research centres in 14 Member States and active patient organizations. The network addresses harmonizing and implementing standards for clinical and diagnostic best practice, improving equity of care provision across Member States, decreasing time to diagnosis, increasing cost efficiency through better care pathways, access to specialist training and education, eHealth services, development and application of care guidelines, facilitating translational and clinical research, harmonising data and samples for research reuse, and data sharing.

EURO-NMD is now an operational network with active participation from our HCPs in all disease and diagnostic working groups. The network has performed a number of gap analyses and has identified priorities for the development or endorsement of guidelines for NMDs. As we move into year two of operation, our priorities are to; endorse or develop guidelines in the areas of highest need, develop an educational training programme, prepare a research plan, deliver the first EURO-NMD training school and consolidate the work of the network. EURO-NMD will begin using the Clinical Patient Management System as the first patients are enrolled EURO-NMD will engage HCP’s in its use to ensure success.


5

P3 Application of NGS technology and establishment of biobanks improved research and diagnostics of rare diseases - Serbian 5 years experience

Pavlovic S, Stojiljkovic M, Tosic N, Zukic B

Institute of Molecular Genetics and Genetic Engineering, University of Belgrade, Belgrade, Serbia

Presenting author: Sonja Pavlovic Contact: [email protected]

Institute of Molecular Genetics and Genetic Engineering has become widely recognized as an expert centre for rare diseases (RD). It is the first institution in Serbia that applied NGS methodology in research and diagnostics of RD. We are also the institution with RD biobank collections containing DNA, RNA, mononuclear cells and tissue samples from over 2000 of patients affected with 50 different diseases.

An accurate diagnosis was provided to over than 100 RD patients who were undiagnosed for years. We used Clinical-Exome Sequencing TruSightOne Gene Panel (4813 clinically-relevant genes), Illumina MiSeq instrument and Illumina VariantStudio. For monogenic diseases, filtration and prioritization of variants were performed according to “in-house” pipeline, using virtual gene panels. Variants were analyzed by various in silico softwares and classified according to ACMG guidelines. Variants selected by these criteria were confirmed by conventional Sanger sequencing and parents’ samples were analyzed whenever available.

Furthermore, novel variants in DNAI1, MUT, PAH, PCCB, SLC37A4, SPAG16 and SPAG17 genes were functionally characterized in adequate in vitro systems such as immortalized patients fibroblasts or Crisper/Cas9 edited commercial cell lines.

Clinical-exome sequencing enabled diagnosis of more than 50 different diagnosis (hematological, metabolic, endocrinological, pulmonary, immunological, orthopedic, dermatological, ophthalmological, cardiological, epileptic encephalopathies etc.). It was particularly important for genetically heterogeneous diseases, such as glycogen storage diseases, branched-chain organic acidurias, primary ciliary dyskinesia, MODY or mitochondriopathies. Moreover, different diseases with overlapping clinical manifestations were accurately diagnosed.

Also, we used TruSeq-Amplicon Cancer Panel to analyse different childhood and adult rare hematological malignancies. Besides studying diagnostic and prognostic malignancy markers, we designed “in-house” virtual pharamocogenomic panel, and performed association studies of pharmacogenomic markers and the course and outcome of rare hematological malignancies, resulting in recommendations for therapeutic modalities in accordance with genomic profile of the patient.

This work has been funded by MESTD, Republic of Serbia (III41004).


P4 RD-Connect Sample Catalogue

van Enckevort DJ1, Wang CM2, Slofstra M1, Hendriksen H1, Reihs R3, Merino Martinez R4, Müller H3, Swertz MA1, Monaco L2

1University Medical Center Groningen, University of Groningen, Groningen, The Netherlands, 2Fondazione Telethon, Milano, Italy, 3Medizinische Universität Graz, Graz, Austria, 4Karolinska Institutet, Stockholm, Sweden

Presenting author: David van Enckevort Contact: [email protected]

The RD-Connect Sample Catalogue was formally released in production in January 2018 (http://samples.rd-connect.eu). It currently shares over seven thousands sample descriptions and a foreseen growth to one hundred thousand samples over the year. It is an important resource for researchers looking for rare disease biological samples relevant to their research. The sample catalogue is tightly coupled with the rest of the RD-Connect infrastructure, especially with the Biobank & Registry Finder, which it queries for the registry and biobank information organisational and contact information.

Current work

Together with the biobanks we are now working on including many more samples in the catalogue. To enable the biobanks to update their own information we have setup a process where the biobank can upload data for their samples in a format that is convenient to deliver, and can be checked in their own private staging area. The sample catalogue will then automatically integrate the information of the different biobanks into one catalogue using the MOLGENIS mapping service, which facilitates and automates the necessary transformations of the data to the common data format of the catalogue.

In collaboration with the BBMRI-ERIC Common Service for IT we will extend the sample catalogue with support for the BBMRI-ERIC Negotiator service, which will allow researchers to make requests for samples directly through the sample catalogue and streamline the process of obtaining research material.

With a select number of biobanks we are performing a pilot with the EUPID system which provides an implementation of the Privacy Preserving Record Linkage principles as outlined by the joint GA4GH and IRDiRC PPRL working group.

Sustainability

The sustainability of the RD-Connect Sample Catalogue is guaranteed by a collaboration with BBMRI-ERIC, who have included Rare Disease research in the 2018 work program, and in part also in the Solve-RD project. Practically this will mean that the Sample Catalogue will be integrated in the BBMRI-ERIC Common Services for IT, which ensures continued hosting and development and will allow to profit from the synergy with the existing tools in the Common Services for IT.


7

P5 The genetic background of non-medullary paediatric thyroid carcinoma (GeNoThyPe)

Van der Tuin K1, Morreau H2, Links TP3, Hes FJ1 and Dutch Paediatric Thyroid Carcinoma Consortium

1Department of Clinical Genetics, Leiden University Medical Centre, Leiden, the Netherlands, 2Department of Pathology, Leiden University Medical Centre, Leiden, the Netherlands, 3Department of Endocrinology, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands

Presenting author: Karin van der Tuin Contact: [email protected]

Non-medullary thyroid carcinoma (TC) is the most frequent endocrine malignancy, representing up to 3% of all malignant tumours in childhood. In general it has a good prognosis; nevertheless in children TC presents at a more advanced stage than in adults. Moreover, it has been found to have relatively high rates of recurrence (7%), persistent disease (8%) and postoperative complications (>30%). The development of a so-called “adult type” of cancer in children is rare and highly suggestive of an underlying genetic predisposition. Also, case control studies have shown a high degree of heritability (3-10 fold increased risk). However, because of the relative rarity of paediatric TC, the frequency of germline mutations in cancer predisposition genes have barely been studied. Our aim is thus to improve the knowledge of the genetic background of paediatric TC by determining the contribution of mutations in known cancer predisposition genes and identifying novel TC susceptibility genes. A close national collaboration has resulted in the assemblage of complete clinical data, tumour tissue and leukocyte DNA for whole genome sequencing from 100 children diagnosed with TC between 1970 and 2017. Our study represents one of the largest cohorts of paediatric TC patients worldwide. For ranking the variants we combine the molecular data with the clinical- and pathological data and focus on three tiers. First, we look for variants in autosomal dominant and recessive cancer predisposition genes. Secondly, we look further into genes involved in pathways of thyroid development, TSH response and tumorigenesis (including micro RNA processing). Thirdly, we combine data of all patients in our cohort to look for genes of which more than one patient has a rare variant. Moreover, we have access to both retrospective and prospective replication cohorts.


P6 The European Genome-phenome Archive: Archiving RD-Connect data

Spalding JD1, The European Genome-phenome Archive2

1European Molecular Biology Laboratory, European Bioinformatics Institute, Wellcome Genome Campus, Hinxton, Cambridge, CB10 1SD, United Kingdom, 2Centre for Genomic Regulation (CRG), Dr. Aiguader 88, 08003 Barcelona, Spain and Universitat Pompeu Fabra (UPF), Barcelona, Spain

Presenting author: Dylan Spalding Contact: [email protected]

The European Genome-phenome Archive (EGA) is a controlled-access archive, which provides secure long-term storage and distribution of human genetic and phenotypic data for biomedical research. The EGA has been archiving the raw data files generated by RD-Connect partners, and then distributing these files for analysis in the Genome-Phenome Analysis Platform. The total volume of files uploaded to the EGA is over 67TB. RD-Connect has been a use case for large scale multi-submitter workflows at EGA, driving developments such as allowing multiple partners to submit on behalf of a single project, improving feedback to submitters and data co-ordinators, and driving developments to link multi-omics submissions correctly within the EGA and also external archives with the aim of making these data FAIR ( Findable, Accessible, Interoperable, and Reusable).

As a driver project for the Global Alliance for Genomics and Health (GA4GH) the EGA aims to use the data submitted to EGA from RD-Connect as a use case for the GA4GH Clinical and Phenotypic Data Capture Work Stream. EGA has also been working with RD-Connect to test the new GA4GH htsget protocol to visualise data archived at EGA submitted by RD-Connect project directly from EGA in genome browsers.

The EGA will continue to work with RD-Connect to ensure all raw data produced by the RD-Connect partners has been archived at EGA, develop SOPs to allow submission of samples with a mixture of controlled and open access attributes, and support the submission of the results of analyses performed by RD-Connect partners and ensure these are correctly linked to the raw data.


9

P7 Playing cards with biobanks: a tool for promoting public engagement

Casareto L1, Baldo C2, Visconti P3, Parodi F3, Parodi B3 and Filocamo M4

1TNGB Coordination Office, IRCCS Istituto Giannina Gaslini, Genoa, Italy, 2S.C. Laboratorio di Genetica Umana, E.O. Ospedali Galliera, Genoa, Italy, 3Centro Risorse Biologiche Ospedale Policlinico San Martino, Genoa, Italy, 4U.O.S.D. Laboratorio Genetica Molecolare e Biobanche, IRCCS Istituto Giannina Gaslini, Genoa, Italy

Presenting author: Lorena Casareto Contact: [email protected]

The assurance of public support is vital to the future sustainability of the biobanks. Therefore the involvement of the citizen in the biobanking is crucial for building and maintaining public trust. Recognising the importance to engage a liaise with public, two RD-biobanks of the Telethon Network of Genetic Biobanks (TNGB) participated in a local initiative organised in collaboration with A.Li.Sa. (Regional Health Authority) and BBMRI.it within the European Biotech Week 2017.

Focus of the event, held in Genoa (Italy) on 26th of September 2017, was the dissemination of the biobanks’ services and the increase of the public awareness. This objective was pursued by developing an interactive ad hoc game card on the main biobank issues, including “recurrent words”, “value”, “main activities”, “actors”, “governance and ELSI”, “alliance”. The game consisted of a deck of cards, each containing multi-choice questions on the main above mentioned biobanking topics with 3 alternative answers of which only one was the correct answer. The cards, gathered by the 6 topics represented by diverse colours, were picked out from the player(s) by rolling a dice. Briefly, the closed-ended questions gave the player the possibility to discuss with the game facilitators (i.e. biobankers) the various biobanking aspects. In addition, a relevant booklet was given to players to foster the spread of biobanking. Thanks to the positive interaction created by the game, the facilitators had the opportunity to discuss with the participants the importance of biobanking and networking in the field of rare disease via presentations and videos concerning some projects such as TNGB and RD-Connect. The event was well-attended by approximately 50 people including citizens, students, health professionals and journalists.


P8 Disease modifiers in Congenital Myasthenic Syndrome: whole genome and transcriptome integration use case

Bosio M1,2, Rigau M1, Esteve-Codina A3, Dabad M3, Azuma Y4, Topf A4, Vazquez M1, Thompson R4, Gut I2,3, Lochmüller H4, Valencia A1,2,5, Beltran S2,3, Laurie S3, 't Hoen PAC6, Capella-Gutierrez S1,2

1 Barcelona Supercomputing Center, Barcelona, Spain, 2Instituto Nacional de Bioinformática, ELIXIR Spain (INB, ELIXIR ES), Spain, 3Centro Nacional de Análisis Genómico-Center for Genomic Regulation (CNAG-CRG), Barcelona Institute of Science and Technology (BIST), 08028 Barcelona, Spain, 4Institute of Genetic Medicine, Central Parkway, Newcastle upon Tyne, NE1 3BZ, UK, 5ICREA Professor, 6Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands

Presenting author: Mattia Bosio, Salvador Capella Gutierrez Contact: [mattia.bosio, salvador.capella]@bsc.es

We processed RNAseq data generated with RD-Connect workflows on patients affected by Congenital myasthenic syndrome (CMS). CMS is a group of conditions characterised by weakness and fatigue resulting from problems at the junction between the nerve and the muscle. CMS can be caused by mutations in different genes, and even individuals with the same mutation can present different degrees of severity and progression. We analysed 20 patients with the same homozygous mutation (delG) in CHRNE, selected across the severity spectrum (8 severe, 10 mild, and 2 moderate). RNA-seq was performed on fibroblast, and WGS from blood, with the goal of identifying modifier variants explaining the different phenotypic expression.

We first attempted to identify a unique source explaining the phenotypic differences, performing differential expression analysis using DESeq2 and correcting for confounding factors. We obtained a list of significantly differentially expressed genes, among which NGFR stood out as a good candidate, being over expressed in 3 of 8 severe cases, and known to be associated with osteoarthritis, a comorbidity present in myasthenic patients. In parallel, we searched for segregating mutations from WGS data, without obtaining conclusive results across the cohort.

We then turned towards a more personalized approach, searching for modifier genes explaining the severity differences. Our hypothesis was that multiple different variants might explain the differences in observed symptoms. We combined differential expression analyses, the detection of mutations, (i.e. CNVs and compound heterozygous), and multi-layer networks associations, finding mutations affecting multiple genes associated with CMS and/or other severe muscular conditions such as HSPG2 and LAMA2.

These results support the use of multiple omics sources to answer complex questions about rare diseases (e.g. what can explain the difference in phenotype among patients with identical causal mutation), highlighting the opportunity to develop customized analytical approaches to complement standard analysis pipelines.


11

P9 RD-Connect Multiomics Working Group: Pilot work integrating WGS and RNASeq pipelines

Bosio M1,2, Rigau M1, Esteve-Codina A3, Dabad M3, Laurie S3, Azuma Y4, Salgado D5, Thompson R4, Topf A4, Gut I2,3, Lochmüller H4, Valencia A1,2,6, Beltran S2,3, 't Hoen PAC7, Capella-Gutierrez S1,2

1Barcelona Supercomputing Center, Barcelona, Spain, 2Instituto Nacional de Bioinformática, ELIXIR Spain (INB, ELIXIR ES), Spain, 3Centro Nacional de Análisis Genómico-Center for Genomic Regulation (CNAG-CRG), Barcelona Institute of Science and Technology (BIST), 08028 Barcelona, Spain, 4Institute of Genetic Medicine, Central Parkway, Newcastle upon Tyne, NE1 3BZ, UK, 5Aix Marseille University, INSERM, GMGF, Marseille, France, 6ICREA Professor, 7Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands.

Presenting author: Mattia Bosio, Salvador Capella Gutierrez Contact: [mattia.bosio, salvador.capella]@bsc.es

Within the RD-Connect consortium we developed a prototype for multiomics data processing and integration. We have initially focused only in genomics data e.g. whole-genome sequencing (WGS) and whole-exome sequencing (WES), and transcriptomics data e.g. RNASeq. We processed each omic data with a different pipeline: applying the state of the art processing pipeline from RD-Connect consortium for WGS/WES, and developing a RNASeq specific pipeline taking as starting point best practices by the ENCODE Consortium and the GATK software toolkit. Moreover, we are constantly improving the workflows to keep the RD-Connect pipeline up with the state of the art.

The main goal is to obtain a reliable set of variant calls per sample within RD-Connect genome phenome analysis platform, integrating and connecting both data sources. As part of this process, a number of output files are produced, which can then be fed to downstream analyses tools (e.g. allele-specific RNA coverage for all variants, gene expression quantification, alignment files, and quality control results). Combining both data sources will contribute to prioritize which genetic variants have a stronger impact on the development of the observed diseases by better characterizing the impact of variants on gene function (e.g. dysregulating expression or splice-altering mutations).

As a pilot test, we processed 20 samples from a cohort of patients affected by Congenital myasthenic syndrome (CMS). For each patient, RNA-seq was performed on fibroblast samples and WGS on blood samples, in order to have both transcriptomics and genomics data. Currently, all samples are processed with both pipelines and VCF data are included in the platform, while the integration process is still undergoing to evaluate the concordance between variant calls from WGS and RNAseq. This result confirms the correct implementation of both pipelines and the integration framework as a useful tool for RD-Connect users.


P10 A comparative study on blood and brain HD signatures: Similarities between mouse and human HD gene expression data

Mina E1, Toonen LJA1, Evers MM2,’t Hoen PAC1, Roos M1, Hettne K1, van Roon-Mom W1

1Department of Human Genetics, Leiden University Medical Center, 2300 RC Leiden, The Netherlands, 2Department of Research & Development, uniQure, Amsterdam, The Netherlands

Presenting author: Eleni Mina Contact: [email protected]

Huntington's disease (HD) is a devastating brain disorder with no effective treatment or cure available. In order to develop interventions that start before disease onset, it is important to have biomarkers that can accurately measure changes between controls and HD patients before symptoms first arise. In a previous study, we identified signatures that were shared between blood and brain tissue at a systems level based on mechanisms that involve multiple genes and proteins. Since many potential drugs are first screened in rodent models we wanted to investigate if these human signatures are also present in the YAC128 HD mouse model.

We sequenced RNA from longitudinal blood samples and end stage brain samples from 8 HD mice and 8 wild type mice. Blood was collected at 12 and 20 months. Four brain regions were examined; brainstem, cerebellum, striatum and cortex.

The blood data analysis showed that the first time point exhibited a very mild pathology with a clear immune response signature, while the second time point exhibited an intermediate pathology. Representative annotations were immune response, autophagy, protein transport, and DNA repair.

Regarding the brain data, cortex and brainstem exhibited a mild pathology while cerebellum and striatum an intermediate to moderate pathology. Representative annotations in common from all four brain regions were immune response, DNA repair, protein transport, chromatin modification and myelination.

The results from the mouse analysis from blood and brain tissue corresponded to the changes previously observed in the human data. In addition, our preliminary results indicate that also in the mouse data, blood exhibits to some extend similarities to the pathology in brain. This is reflected by the processes immune response, DNA repair and protein transport. Our next step is to perform a computational randomization experiment to determine statistically the similarities between the blood and brain signatures in mouse.


13

P11 Developing the international EuroBioBank network of rare diseases biobanks within the RD-Connect platform

Wang CM1, Benvenuti S1, Monaco L1

1Fondazione Telethon, Milan, Italy

Presenting author: Mary Wang Contact: [email protected]

Access to high quality human biological materials is fundamental for research in rare diseases (RD). It underpins the understanding of RD pathogenesis, development of diagnostic tools, identification of therapeutic targets and validation. The RD-Connect biobank pillar engages expertise of biobank managers, IT developers and network coordinators across Europe to develop an international network of rare diseases biobanks that is supported by RD-Connect tools.

The RD-Connect biobank network is EuroBioBank. Founded in 2001, EuroBioBank is currently a network of 25 RD biobanks located in 11 countries. It stores more than 148,000 biospecimens for research, covering more than 950 RDs. EuroBioBank biobanks are quality biobanks which have all underwent an assessment ensuring they meet the minimum quality standards. The biobanks are listed in the RD-Connect Registry & Biobank Finder (catalogue.rd-connect.eu) and their sample collections are listed in the RD-Connect Sample Catalogue (samples.rd-connect.eu).

New RD biobanks are invited to join EuroBioBank, and the application is managed through a dedicated web interface within the RD-Connect Registry & Biobank Finder. RD biobank can submit its membership application by completing a questionnaire on its biobanking activities, which is reviewed by a panel of experts. The Panel covers expertise on biobanking operations, quality, IT infrastructure capability and ethical, legal and social issues (ELSI). Member biobank can increase its visibility and bring positive impact to the RD scientific community by linking biological samples, via RD-Connect Sample Catalogue to correlated clinical and omics data within the RD-Connect platform.

In addition to facilitating sample access, RD-Connect EuroBioBank promotes the maximum exploitation of valuable RD resources and data, following the findable, accessible, interoperable and re-usable (FAIR) data principles. Continued strategic decisions and partnerships have allowed EuroBioBank to evolve and be equipped to support basic and translational research in rare diseases.


P12 FAIRification of rare disease patient registries

Jacobsen A1, Thompson M1, Sernadela P2, Ehrhart F3,4, Waagmeester A5, Carta C6, Groenen K7, Kersloot MG8,9, Kool LS7, Mordenti M10, Locatelli M10, Sangiorgi L10, Schultes EA11, Cornet R8, Wilkinson MD12, Taruscio D6, Thompson R13, van Enckevort D14, Evelo CT3,4, Roos M1

1Department of Human Genetics, Leiden University Medical Center, Leiden, The Netherlands, 2Universidade de Aveiro, Aveiro, Portugal, 3Department of Bioinformatics - BiGCaT, NUTRIM, Maastricht University, The Netherlands, 4GKC-Rett Expertise Centre, Maastricht University Medical Center, The Netherlands, 5Micelio, Antwerp - Ekeren, Belgium, 6National Centre for Rare Diseases, Istituto Superiore di Sanità, Rome, Italy, 7Department of Radiology and Nuclear Medicine, Radboud University Medical Centre, Nijmegen, The Netherlands, 8Academic Medical Center, dept of Medical Informatics, Amsterdam, The Netherlands, 9Castor EDC, Amsterdam, The Netherlands, 10Medical Genetic Department & CLIBI Lab, Rizzoli Orthopaedic Institute, Bologna, Italy, 11GO FAIR International Support and Coordination Office, Leiden, The Netherlands, 12Universidad Politécnica de Madrid, Spain, 13Newcastle University, Newcastle upon Tyne, The United Kingdom, 14University of Groningen, University Medical Center Groningen, Groningen, The Netherlands

Presenting author: Annika Jacobsen Contact: [email protected]

Rare diseases (RDs) have low prevalence, frequently cause severe conditions, and the molecular causes are often not very well known, making diagnosis and treatment difficult. Patient registries, however, contain crucial information to advance our knowledge of RDs, and it is therefore in the patients’ interest that the data is made FAIR (Findable, Accessible, Interoperable, and Reusable for humans and computers) at the source. Interoperability is especially important for RD data as they often are sensitive, sparse, highly distributed and heterogeneous. In RD-Connect, we make data interoperable by describing them in terms of ontologies, which clarifies how data elements are related, also for computers. The degree of interoperability is greatly impacted by the choice of ontologies and the availability of ontology mapping tools and mappings. Therefore, to ensure optimal interoperability between RD data, we need guidelines to streamline our choices.

Here, we describe our best practices with FAIRifying data of selected rare diseases such as: Duchenne Muscular Dystrophy, Vascular Anomalies, Rett syndrome, and Osteogenesis Imperfecta. Doing this requires international interdisciplinary collaboration of disease experts, patient representatives, FAIR data experts, and software engineers. The FAIRification procedure consists of the following steps: 1) set the scope of your FAIRification process, i.e. achieve increased interoperability or a clearer definition of accessibility, 2) create machine readable metadata, 3) create machine readable data by i.e. using specific ontologies such as Human Phenotype Ontology and the Orphanet Rare Disease Ontology (IRDiRC recognized resources), 4) depositing machine readable data and metadata on a FAIR data point, and 5) explore added benefits of your FAIR data, by i.e. answering specific cross-resource questions.

A detailed description for each of these steps is provided for other researchers to make their RD data FAIR, thus stimulating convergence within the RD domain.


15

P13 Modularizing RD-Connect Central Authentication Service components into software containers

Fernández JM1,2, Cañada A3, Piscia D4, Dawson J5, Beltrán S4, Valencia A1,2,6, Capella-Gutierrez S1,2

1Barcelona Supercomputing Center (BSC-CNS) , Barcelona, Spain, 2Instituto Nacional de Bioinformática, ELIXIR Spain (INB, ELIXIR ES), Spain, 3Centro Nacional de Investigaciones Oncológicas (CNIO), Spain, 4Centro Nacional de Análisis Genómico (CNAG-CRG), Center for Genomic Regulation, Barcelona, Spain, 5Institute of Genetic Medicine, Newcastle University, United Kingdom, 6ICREA Professor

Presenting author: José M. Fernández Contact: [email protected]

The RD-Connect platform, with more than 200 users, connects and integrates sensible human data, rare disease developments and knowledge including medical records from several registries, sample metadata from biobanks, molecular data, primary and secondary analysis data and associated bioinformatics resources used to process, analyze and integrate them. All these resources have their specific, independent web platforms, and most of them have some kind of fine-grained authorization mechanisms in order to restrict the access to sensible information. Although these control mechanisms are managed independently, an infrastructure which allows creating a secure environment among the different scientific resources is mandatory in order to use them in the RD-Connect platform.

The RD-Connect Central Authentication Service (CAS) components have been setup and deployed to play this role, providing a single sign-on facility trusted by all these distributed resources. Although all the components can run in the same host, as each one of these components are playing a critical mission, we have focused our efforts in modularizing them and deploying them into software containers, so they can be improved, maintained, updated and redeployed independently. Importantly, this modular design and implementation would facilitate their reuse and extension in future projects. For instance, the interconnection with the ELIXIR ecosystem is currently underway thanks to the incorporation of the ELIXIR AAI (Authentication and Authorisation Infrastructure).


P14 ALFA v3: ALamut Functional Annotations and Predictions

Blavier A1, Palvadeau R1, Hatzoglou A1, Mancini F1

1Interactive Biosoftware, Rouen, France

Presenting author: André Blavier Contact: [email protected]

Interactive Biosoftware contributes to the RD-Connect global infrastructure and analysis platform by developing a gene regulation prediction system, named Alamut Functional Annotations (ALFA), to identify genetic variations located in non-coding DNA that may be involved in gene regulation processes. Genetic variations that do not directly impact the protein sequence are quite diversified in their genomic locations, and may affect almost all processes of gene regulation from transcription to post-translation. ALFA, the ALamut Functional Annotations system, stores proximal and distal regulatory regions annotations (such as promoters, transcription factor binding sites, enhancers, microRNA target sites) along with cell type or tissue specificity gathered from active international projects such as Ensembl Regulatory or FANTOM5. ALFA allows users to explore a large number of functional elements in non-coding regulatory regions of the Human genome (GRCh37 and GRCh38 are both supported) providing opportunities to identify and investigate non-coding regulatory variants. ALFA offers a REST API and a command-line interface to make its integration into the RD-Connect platform easier for bioinformaticians. ALFA also offers a graphical web interface to help researchers detect and annotate regulatory variants that may have functional effect on gene expression. Based on the ALFA database and both allele frequency and classified variations databases Interactive Biosoftware is now exploring ways to apply machine-learning techniques in order to predict potential effects of variants on gene regulation and expression.


17

P15 Training activities to promote, create, and support International and FAIR rare disease registries

Carta C1, Roos M2, Torreri P1, Jacobsen A2, van Enckevort D3, Kodra Y1, Cornet R4, Kaliyaperumal R2, Thompson M2, Via A5, Licata L6, Agresta L1, Crialese P1, Taruscio D1

1Centro Nazionale Malattie Rare, Istituto Superiore di Sanità, Rome, Italy, 2 Department of Human Genetics, Leiden University Medical Center, Leiden, The Netherlands, 3University of Groningen, University Medical Center Groningen, Groningen, The Netherlands, 4Academic Medical Center, dept of Medical Informatics, Amsterdam, The Netherlands, 5National Research Council, Rome, Italy, 6University of Rome Tor Vergata, Rome, Italy

Presenting author: Claudio Carta Contact: [email protected]

Over the last six years a growing number of courses have been organised in order to promote, create, and support International and FAIR (Findable, Accessible, Interoperable, and Reusable for humans and computers) rare disease registries with consequent production of training materials.

Centro Nazionale Malattie Rare, Istituto Superiore di Sanità (CNMR-ISS), Rome Italy is actively involved in these training activities. Here we present two examples of courses organised and hosted by CNMR-ISS: “The International Summer School on Rare Disease and Orphan Drug Registries” and “Bring Your Own Data, BYOD, to link rare disease registries”.

They have consistently evolved over the years thanks to the successful collaboration and support from the key partners including RD-Connect, the linked data and ontology task force, ELIXIR (in particular DTL/ELIXIR-NL and ELIXIR-IIB), LUMC, UMCG, EURORDIS, ERNs, EpiRare, RD-Action, and ICORD.

In the first course, the didactic methodology used has become the highly interactive One Day-One Problem Problem-based Learning in which participants, in small groups guided by a Facilitator and with the support of experts, learn how to solve pre-defined problems. While in the BYOD, where the participants get further into the FAIRification process, new elements have been added over the years for example, a time slot for a semantic model sketching exercise.

This event, the vast number of materials, and tools produced have supported the development of a FAIR rare disease registries. This is proven, for example, by the growing number of accepted abstracts, at international conferences, over the last six years.

Since 2018 these courses, started in different years for diverse audiences, have become one unique event.


P16 Recommendations for improving the quality of rare disease registries

Kodra Y1, Weinbach J2, Coi A3, Posada-de-la-Paz M4 , Santoro M3, Lemonnier L5, Van Meel M9, David van Enckevort6, Marco Roos7, Annika Jacobsen7, Ronald Cornet8 , Virginie Bros-Facer9, S. Faisal A10 , Daniel Renault D9 , Rainald von Gizycki R9, Popa V9, Landais P11, Bianchi F3, Torreri P1, Carta C1 , Mascalzoni D12 , Gainotti S13 , Lopez E4 , Heimo Müller H14, Reis R14, Ambrosini A15, R. Rubinstein Y16, Lochmüller H17 and Taruscio D1

1National Centre for Rare Diseases, Istituto Superiore di Sanità, Rome, Italy, 2RaDiCo, The French National Programme on Rare Disease Cohorts. National Institute of Health and Medical (3) Research, UMR S933, Paris, France. www.radico.fr, 3Institute of Clinical Physiology, National Research Council, Pisa, Italy; Fondazione Toscana “Gabriele Monasterio” (FTGM), Pisa, Italy), 4Institute of Rare Diseases Research, ISCIII, Spain, RDR and CIBERER, Madrid, Spain, 5Vaincre la Mucoviscidose, the French Cystic Fibrosis Patient Organization, Paris, France, 6Department of Genetics, University Medical Centre Groningen (UMCG) Nederlands7Leiden University Medical Center, Leiden, The Netherlands, 8Academic Medical Center, dept of Amsterdam Public Health research institute Medical Informatics, Amsterdam The Netherlands, 9Patient Advisory Council of RD-Connect coordinated by EURORDIS-Rare Diseases Europe, 10Office For Rare Conditions, Royal Hospital for Children, University of Glasgow, United Kingdom, 11EA2415 Clinical Research Institute, Montpellier University, France The French National Programme on Rare Disease Cohorts. National Institute of Health and Medical Research, UMR S933, Hôpital Trousseau, Paris, France. www.radico.fr, 12Uppsala University, Sweden, 13Bioethics Unit, Office of the President, Istituto Superiore di Sanità, Rome, Italy, 14Medical University of Graz, Vienna, 15Fondazione Telethon, Milan, Italy, 16National Library of Medicine/National Institutes of Health, Bethesda, MD, USA, 17Department of Neuropediatrics and Muscle Disorders, Medical Center – University of Freiburg, Faculty of Medicine, Freiburg, Germany; Centro Nacional de Análisis Genómico, Spain

Presenting author: Yllka Kodra Contact: [email protected]

In the field of rare diseases (RD), patient registries are powerful resources for developing clinical research, to facilitate the planning of appropriate clinical trials, to improve patient care and support healthcare management. The importance of registries, databases and information networks for RD has been recognized also at European level in the document “EU Council Recommendation of 8 June 2009 on an action in the field of rare diseases”. Patient registries constitute a key information system to support the forthcoming activities of European Reference Networks (ERNs) on RD. A rapid proliferation of RD registries has occurred over the last years and significant variability exists in their structure, aims, data content, quality, and governance. In response to these heterogeneities, in the framework of RD-Connect (http://rd-connect.eu/), the National Center of Rare Diseases in Italy coordinated and selected a group of experts including members of patients organizations, with wide experience in the field of RD registries. The experts are included as authors of this article. The objective of the expert working group was to make a list of recommendations for improving the quality of RD registries that would serve as a framework for the future.

Quality of registries is a global concern and involves all sequential activities for running a registry: establishment of the registry governance, identification of the right data sources, specification of Data Elements and standardized Case Report Form (CRF), construction of a secure-by-design Central Database Infrastructure, production of data quality and to the dissemination of a quality information in terms of health statistics, outcomes, scientific publishing. Other topics such as developing adequate documentation, training of staff and providing data quality audit are also considered as essential for improving registry quality. The experts are working on the development of recommendations, essential for empowering existing and new registry operators, emphasizing the importance of maintaining a high-quality registry.

http://rd-connect.eu/

http://rd-connect.eu/


19

P17 Detecting consanguinity based on whole-exome sequencing: integration within the RD-connect genome-phenome analysis platform and impact on the molecular diagnosis of rare disease patients

Papakonstantinou A1,2, Matalonga L1,2, Laurie S1,2 , Piscia D1,2 ,Beltran S1,2

1CNAG-CRG, Center for Genomic Regulation (CRG), Barcelona Institute of Science and Technology (BIST), Baldiri i Reixac 4, 08028 Barcelona, Spain, 2Universitat Pompeu Fabra (UPF), Barcelona, Spain

Presenting author: Anastasios Papakonstantinou Contact: [email protected]

Inbreeding is known to be associated with an increased risk of rare diseases. NGS technologies enable the detection of runs of homozygosity (ROH), which are genomic regions where a reduction in heterozygosity is observed due to identity by descent. This provides the opportunity to estimate consanguinity at the level of an individual's genome or exome. In this study, we present the benefits of combining WES and ROH analysis within the RD-connect Genome-Phenome Analysis Platform for the filtering and interpretation of genomic data.

Homozygosity mapping was performed for 2,198 individuals from different projects that have been shared in RD-Connect for which both clinical and genomic data was available. ROH were estimated using an optimised protocol for WES data in PLINK and the effect of different clinical and genomic parameters such as sequencing strategy, quality filtering, ethnicity and family history was assessed. Filtering by ROH was integrated into the analysis platform and WES data were analysed using the RD-connect interface.

The number and sum length of ROH were computed for each individual. Results were compared with the consanguinity status reported in PhenoTips and a threshold defined according to which an individual can be classified as likely to be the progeny of a consanguineous pairing or not. This approach showed high specificity and sensitivity when applied for homozygosity detection in individuals from an independent project (URDCAT). Filtering by ROH regions in the genome-phenome analysis platform assisted in the identification of pathogenic or candidate causative variants in 52% of cases that were determined to be consanguineous according to either their clinical record or the ROH analysis.

In this study, a threshold to estimate consanguinity based on WES has been established. Furthermore, the integration of filtering by ROH within the RD-connect analysis platform has been shown to facilitate the molecular diagnosis of rare disease patients.


P18 Challenges and successes of sharing data for resolution of rare disease cases: The example of the BBMRI-LPC WES call, in collaboration with EuroBioBank and RD-Connect

Laurie, S1,2, Beltran, S1,2, Bayes, M1,2, Dawson, J3, Fusté, B1,2, Gut, M1,2, Lochmüller, H3, López-Martín, E4,5, Matalonga, L1,2, Monaco, L6, van Ommen, G7, Piscia, D1,2, Posada, M4,5, Sims, S8, Thompson, R3, Wang, M6, Zeggini, E8, Gut, I1,2, BBMRI-LPC Consortium

1Centro Nacional de Análisis Genómico (CNAG-CRG), Barcelona, Spain, 2Universitat Pompeu Fabra (UPF), Barcelona, Spain, 3Institute of Genetic Medicine, MRC Centre for Neuromuscular Diseases, Newcastle University, Newcastle, UK, 4Institute of Rare Diseases Research, IIER-ISCIII, Madrid, Spain, 5Centre for Biomedical Network Research on Rare Diseases, CIBERER, Madrid, Spain, 6Fondazione Telethon, Milan, Italy, 7Department of Human Genetics, Leiden University Medical Center, Leiden, The Netherlands, 8Wellcome Trust Sanger Institute (WTSI), Hinxton, UK

Presenting author: Steve Laurie Contact: [email protected]

Sharing of data, both genetic and phenotypic, is essential for facilitating molecular diagnoses of rare disease cases. However, the act of sharing remains challenging for technical, economic, political, and cultural reasons. The BBMRI-LPC WES call aimed to promote such sharing, by offering a unique free-of-charge opportunity to genetically diagnose rare disease patients with biological samples deposited within the EuroBioBank network. 800 whole exomes were sequenced from 17 distinct projects, each having 2-3 principal investigators from different countries. The projects spanned a wide range of rare disease phenotypes, and informed consent had to permit data sharing for research purposes through controlled access repositories such as the EGA (https://ega.crg.eu/) and RD-Connect (https://platform.rd-connect.). Clinical and phenotypic information for every case was collected in RD-Connect’s customised PhenoTips instance, using standards such as the Human Phenotype Ontology (HPO), OMIM and Orpha codes, and sequencing undertaken jointly at the CNAG in Spain, and the WTSI in the UK. Sequencing data was processed using the RD-Connect standard analysis pipeline and results made available through the RD-Connect Genome-Phenome Analysis Platform once all call requirements had been met. The Platform allows researchers to analyse and interpret their genotype:phenotype data privately for up to 6 months before it is shared with other authorised users, and also facilitates anonymised data sharing through APIs from initiatives such as the GA4GH/IRDiRC MatchMaker Exchange. We will report on the challenges and lessons learned from conducting such a complex transnational collaborative initiative and present an up-to-date diagnostic yield of the project as a whole while focusing on some success stories.


21

P19 The RD-Connect Registry & Biobank Finder: an online directory for sharing and integrating RD data and samples from RD registries and biobanks

Torreri P1, Gainotti S1, Carta C1, Kodra Y1, Wang CM2, Monaco L2, Reihs R3, Mueller H3,Taruscio D1

1Istituto Superiore di Sanità, Rome, Italy, 2 Fondazione Telethon, Milan, Italy, 3Medical University of Graz, Graz, Austria

Presenting author: Paola Torreri Contact: [email protected]

RD research should be based on the systematic collection and reuse of available data which should be Findable, Accessible, Interoperable and Reusable (FAIR). However still too often RD data, including registry data and data on bioresources availability, are collected and stored in siloed databases with little information on their content and possibility to access them. The RD-Connect Registry & Biobank Finder aims to concentrate sparse information on RD patients in one unique source by showing the number of samples/cases included in each databases, the possibility to access the data, and the integration of data among biobanks and registries. Each database has a dedicated ID-Card which consists in a webpage with information including: contact details, diseases collected with aggregated data and metadata data such as standard operating procedures, documents for the informed consent, case report forms and other relevant documents. Importantly, the “disease matrix” (DM) section provides information on the number of patient cases and biological samples held for each RD, with its associated Orphacode, OMIM and ICD10 codes. Currently Registry & Biobank Finder is populated with aggregated data for more than 230 registries and 21 biobanks of the EuroBioBank network.

The functioning and results of Registry & Biobank Finder, have recently been published and video tutorials are now available to explain how to navigate the system and how to create an ID-card for registries and biobanks. Registries from all countries worldwide are participating, with a strong contribution from US (24,8%), followed by International registries (18%) Italian (12,6%), French (7,7%), German (5,9%), UK (5,4%) and Spanish registries (4,5%) and most of them belong to a specific network. The most represented disease categories in RD registries are rare neurological diseases (17,1%), rare neuromuscular diseases (14%), rare hematological diseases (11,3%), rare pulmonary diseases (9,5%) rare renal diseases (8,1%) and rare hereditary metabolic disorders (7,2%) National programs or platforms. The classification of ID-Cards into disease areas based on the European Reference Networks (ERNs) has underlined the potential role of the Registry & Biobank Finder within the ERN community in order to increase the visibility of the registry activities within and to become an important tool according the European Union need in the registries field.


P20 Mutation in POLR3K cause hypomyelinating leukodystrophy and abnormal ribosomal RNA regulation

Dorboz I1, Dumay-Odelot H2,3,Boussaid K4, Bouyacoub Y5, Barreau P1,Samaan S1,6, Jmel H5, Eymard-Pierre E7, Cances C8, Bar C8, Desportes V9, Renaldo F4, Elmaleh-Bergès M10,Teichmann M2,3, Boespflug-Tanguy O1,4

1INSERM UMR 1141 PROTECT Université Paris Diderot- Sorbonne Paris Cité, France, 2Université de Bordeaux , ARNA Laboratory , F-33076 Bordeaux , France, 3INSERM, U1212 - CNRS UMR 5320 , ARNA Laboratory , F-33000 Bordeaux , France, 4Service de Neurologie Pédiatrique et Maladies Métaboliques, Centre de référence des leucodystrophies et leucoencéphalopathies de cause rare (LEUKOFRANCE), CHU APHP Robert-Debré, Paris, France, 5LR11IPT05, Biomedical Genomics and Oncogenetics Laboratory, Institut Pasteur de Tunis - Tunis – Tunisie, 6Department of Medical Genetics, UF Molecular Genetics, CHU APHP, Robert-Debré Paris, France, 7Service de Cytogénétique Médicale, CHU Clermont-Ferrand, France, 8Unité de Neurologie Pédiatrique, CHU Hôpital des Enfants - Toulouse – France, 9Hôpital Femme Mère Enfant, Neurologie pédiatrique, Hospices Civils de Lyon, Bron, France, 10Department of Pediatric Radiology, CHU APHP Robert-Debré, Paris, France

Presenting author: Imen Dorboz Contact: [email protected]

RNA polymerase III (Pol III)-related leukodystrophies belong to the group of autosomal recessive hypomyelinating leukodystrophies characterized by a progressive cerebellar spastic syndrome with variable age of onset and associated non neurological signs (hypo/oligodontia, hypogonadotropic hypogonadism). Mutations have been reported in genes coding for two Pol III specific subunits, POLR3A, POLR3B and one common to Pol III and Pol I complexes, POLR1C. Here we report for the first time a homozygous mutation in a gene coding for another Pol III specific subunit, POLR3K (c.121C>T/(p.Arg41Trp), in two unrelated patients. Our patients expressed neurological and extraneurological signs classically found in Pol III related leukodystrophies with a severe digestive dysfunction not yet reported. The mutation impaired the POLR3K-POLR3B interactions resulting in zebrafish in abnormal gut development. Functional studies in the two patients’ fibroblasts revealed a severe decrease (60-80%) in the expression of ribosomal 5S and 7SL RNAs in comparison with control. These observations underlined the key role of translation regulation in the development and maintenance of the white matter and the cerebellum as already reported for diseases related to genes involved in transfer RNA or translation initiation factors.


23

P21 RD-Connect engagement in Central Eastern and Central Europe

Macek M

Department of Biology and Medical Genetics, Charles University – 2nd Faculty of Medicine and Motol University Hospital, Prague, Czech Rep.

Presenting author: Milan Macek Contact: [email protected]

RD-Connect has participated in numerous conferences in Central and Eastern Europe (WP7) and presented an overview of project activities, with as particular focus on the RD Connect Genome-Phenome Analysis Platform, including presenting the role of registries and biobanks. We attempted to link scattered and often underdeveloped rare diseases activities to major European Union initiatives, including Orphanet. We have also fostered participation of colleagues from this region in the European Reference Networks on Rare Diseases. Overall, RD-connect tried to target key regions and participated in regional / national conferences. Initially, a Central and Eastern European “Chafea Conference of Rare Diseases” organized in Prague, Czech Republic (2014) which set the stage and attracted over 250 participants. RD-connect was present at Georgian Pediatrics meeting in Tbilisi, Georgia (2015), Balkan Congress of Human Genetics in Belgrade, Serbia (2015) and organized the “IRDiRC workshop on Rare genetic diseases, diagnosis and discovery” in Prague, Czech Republic (2014). In addition, RD-connect was present at genetics / rare disease conferences such as in Athens, Greece (2016); Bratislava, Slovakia (2016, 2017); Sofia, Bulgaria (2017); Lviv and Kiev, Ukraine (2017, 2018); Vilnius – Lithuania (2017, 2018); St. Petersburg and Moscow, Russia (2016, 2017); Bucharest and Cluj-Napoca, Romania (2016-2017) and in Krakow, Poland (2017). Finally, RD-connect in collaboration with RD-Action project was co-organising the national Europlan Workshop in 2017 in Prague, which attracted broad participation from Central and Eastern Europe. All these activities markedly contributed to the improvement of research and diagnostics of rare diseases within this target region and promoted the activities of the RD-connect project.


P22 electronic Pharmacogenomics Assistant (ePGA)

George P. Patrinos1, Kleanthi Lakiotaki2, Evgenia Kartsaki2, George Potamias2, Theodora Katsila1, Alexandros Kanterakis2

1University of Patras, School of Health Sciences, Department of Pharmacy, University Campus, Rion, 26504, Patras, Greece; 2Institute of Computer Science, Foundation for Research & Technology-Hellas (FORTH), N. Plastira 100, Vassilika Vouton, 70013 Heraklion, Crete, Greece

Presenting author: George P. Patrinos Contact: [email protected]

The electronic Pharmacogenomics Assistant (ePGA; www.epga.gr) has been developed as a web-based system that offers two main services to the biomedical community involved in pharmacogenomics, namely: (i) explore – a service to search and browse through established pharmacogenomic gene-variant-drug-metabolizer status associations, and (ii) translate – a service to infer metabolizing phenotypes from individual genotype profiles for all known pharmacogenes. ePGA can be of interest to health professionals and biomedical researchers, in addition to the general public that has a keen interest in human genetics. Preliminary results are highly predictive and indicate the potential of the envisaged tool as well as its impact in the clinical application of pharmacogenomics, particularly as soon as this tool gets fully integrated into the battery of the RD-Connect tools. Lastly, ePGA is equipped with an API than allows 3rd party data access. It also employs an ORCID based authentication mechanism with which users can submit novel PGx information, using the microattribution approach.

The research leading to these results has received funding from the European Community’s Seventh Framework Programme (FP7/2007-2013) under grant agreement FP7-2012-HEALTH, “RD-CONNECT” (305444), An integrated platform connecting databases, registries, biobanks and clinical bioinformatics for rare disease research


25

P23 Aix-Marseille University RD-Connect contribution

Salgado D1, Desvignes JP1, Garibal M1, Ghadi R1, Blandin G1, Guien C1, Blanchard A1, Miltgen M1, Pinard A1,

Collod-Béroud G1, Béroud C1,2

1Aix Marseille Univ, INSERM, MMG, Marseille, France, 2APHM, Hôpital TIMONE Enfants, Laboratoire de Génétique Moléculaire, Marseille 13385, France

Presenting author: David Salgado Contact: [email protected]

During the RD-Connect project, the Aix-Marseille University (AMU) partner was the WP4 "clinical bioinformatics" leader. As such it actively contributed to many parts of this ambitious EU funded project:

Creation and development of bioinformatics systems to handle and analyze NGS data: i) The UMD-predictor system to predict the impact of Human cDNA substitutions was published in Human Mutation in 2016 and cited 24 times; ii) The Human Splicing Finder v3, which is the reference system (more than 1000 citations) to predict the impact of intronic or exonic mutations on splicing signals; iii) The VarAFT system to annotate and prioritize variations from NGS experiments (under review).

AMU also created three systems to facilitate the design of therapeutical strategies: i) Skip-E for exon-skipping through Antisense OligoNucleotides (AON); ii) The Crawfish system to design transsplicing molecules (3’ and 5’ SMART) to correct most human mutations and, iii) The NR-Analyzer system to identify nonsense mutations eligible for nonsense readthrough therapies.

AMU also participate to various WP2 deliverables linked to patients registries such mapping RD registries, define common data elements and the registries' catalogue.

AMU was also an important partner for the RD-Connect platform by organizing RD-Connect jamborees to provide guidance in the development of this invaluable resource. AMU shared its systems as VarAFT to build the platform interface. It led the Standard Analysis pipeline Task force that produced the platform's analysis pipeline. In addition, the AMU systems were made available through the platform. Finally, as AMU is now leading the new Elixir Human CNV community, it will also lead the CNV data analysis Task force to implement this critical feature in the platform.

AMU organized and participated to various trainings to promote RD-Connect at the international level (ELIXIR, HVP, 3Gb-Test, HGVS, publications …) and stimulate collaborations (EHR4CR, RD-Sympati …).


P24 F2Share: Federation Framework for sharing research data

Merino-Martinez R1, Amatya S1, Müller H2, Reihs R2

1Karolinska Institutet, 2Medical University of Graz

Presenting author: Roxana Merino-Martinez Contact: [email protected]

F2Share is a generic and semantic agnostic tool to federate data from heterogeneous data sources. Each federation defines its own standard data model that is used to harmonise the shareable data from the members of the federation. Data for sharing is reflected in real-time on a web visualization application. The fundamental software components of this framework are based on the open-source software “Elastic Stack” ensuring sustainable maintenance of the framework. The in-house developed software modules for integration and mapping of data are freely available and documented on GitHub.

Due to the abstract architectural design of this software framework, data from different sources and with different semantics, can be easily integrated with a minimum of effort from data providers. For instance, data from biobanks, research studies, clinical data, bioinformatics analysis metadata, etc. can be integrated, visualized and shared. Each member of the federation will: (1) Map its data to the standard established by the federation, (2) allocate a data repository to export the data for sharing, (3) install a light software to notify when the data in the repository has changed. The central server of the federation extracts, transforms and loads data from each member in a central index from where data is visualized through a flexible and state-of-the-art visualization and exploration tool.

In RD-Connect, F2Share can be used to federate biobank data directly from the source (biobanks). Data from the federation can be exported systematically to be imported into the sample catalogue or the sample catalogue can use the ES APIs to query the indexes. Due to the abstract concept behind F2Share, it is even possible to integrate metadata from registries, bioinformatics, clinical data, other federations, etc.

Documentation:

https://github.com/MIABIS/miabis-mapper

https://github.com/MIABIS/miabis-mapper/wiki

https://github.com/MIABIS/logstash-configuration-generator

https://www.elastic.co

Examples:

Mapper: http://bbmri-biobank.meb.ki.se/

Federated sample data: http://bcnet.meb.ki.se/goto/c4bccfda540269b75a8c8cce4341bb47

https://github.com/MIABIS/miabis-mapper

https://github.com/MIABIS/miabis-mapper/wiki

https://github.com/MIABIS/logstash-configuration-generator

https://www.elastic.co/

http://bbmri-biobank.meb.ki.se/

http://bcnet.meb.ki.se/goto/c4bccfda540269b75a8c8cce4341bb47


27

P25 Collaboration in NeurOmics: enabling effective data-sharing and maximising impact in neuromuscular disease

Turner C1, Straub V1, Thompson R1, Riess O2, van Ommen GJ3; Wirth B4

1John Walton Muscular Dystrophy Research Centre, Newcastle University, UK, 2Tübingen University, Germany, 3Leiden University Medical Center, Netherlands, 4University Hospital Cologne, Germany

Presenting author: Cathy Turner Contact: [email protected]

NeurOmics used the most sophisticated -omics technologies with the aim of revolutionising diagnostics and developing pathomechanism-based treatments for ten rare neuromuscular and neurodegenerative diseases. The approach relied on wide, secure and effective data sharing to maximise impact on disease understanding and therapy development. The project ended in October 2017 but worked closely with RD-Connect on data-sharing policies and agreements and partners continue to make use of the RD-Connect platform. This model of successful data-sharing in order to promote collaboration and accelerate research could be applied across the rare disease field and is summarised here.


P26 The Malta BioBank (BBMRI.mt) in RD-Connect

Vella J, Soler D, Scerri CA, Vella N, Aquilina J, Borg I, Said E, Spiteri A, Dalli J, Ryabova L, Pace M, Borg J, Grech L, Camilleri A, Zammit E, Said-Conti V, Pace N, Vassallo J, Felice AE

Centre of Molecular Medicine and Biobanking, University of Malta, Malta

Presenting author: Joanna Vella Contact: [email protected]

The Malta BioBank (BBMRI.mt); Centre of Molecular Medicine and Biobanking, University of Malta is a founding partner in RD-Connect. It is registered on the RD-Connect Registry and Biobank Finder. More details in www.um.edu.mt/biobank.

BBMRI.mt participated in the BBMRI-LPC whole exome sequencing (WES) call with two collaborative research projects: undiagnosed mitochondrial disorders and undiagnosed neuropathies. WES and bioinformatics analysis were conducted at CNAG-CRG. Phenotypic data were recorded on PhenoTips. Exome data was analysed on the RD-Connect Genome-Phenome Analysis Platform. Rare missense mutations in the mitochondrial cytochrome B gene (MT-CYB) at positions 14766 and 15326 were present in 6 and 11 probands respectively.

Biobank-led research in globin gene disorders focused on strategies to re-activate foetal haemoglobin to therapeutic levels in thalassaemia patients. DNA variants in the human Kruppel Like-Factor 1 (KLF1) gene and others that regulate gamma- to beta globin gene switching were characterised. 420 subjects with HbA2 between 3.1 and 4.5% were retrospectively collected from the biobank. More than half had a mutation in KLF1.

An increasing number of studies point to genetic changes as the cause of congenital renal malformations. 10 patient samples were subjected to whole-genome sequencing. The coding and splice-site regions of 96 genes implicated in non-syndromic CAKUT and involved in kidney development are being studied.

Another project focuses on rare monogenic subtypes of diabetes e.g. MODY. 30 probands and their families with suspected MODY were recruited. Preliminary analysis has identified rare and novel mutations in several genes, including GCK, HNF1α and HNF1β, WFS1 with evidence for several founder effects.

BBMRI.mt was instrumental in placing Rare Diseases (RD) on the agenda of Malta’s EU Presidency. BBMRI.mt works closely with the National Alliance for Rare Diseases Support – Malta to create awareness. The biobank is seeking to reach into community health professionals to strengthen Stakeholder Engagement, Education and Dialogue (SEED).


29

P27 Recommended Practices for Data Standardisation in the Context of the operation of European Reference Networks

Hedley V1, Kohler S2, Roos M3, van Enckevort D4, Rath A5

1Institute of Genetic Medicine, Newcastle University UK; 2Charité, Germany; 3LUMC, Netherlands; 4UMCG, Netherlands; 5Orphanet/INSERM, France

Presenting author: Victoria Hedley Contact: [email protected]

Since 2015, the Policy & Integration workpackage of RD-ACTION (European Joint Action for Rare Diseases) has prioritised support for the creation and implementation of European Reference Networks (ERNs). At the heart of the ERN concept is the principle that wherever possible (and appropriate), expertise will travel rather than the patients themselves. In practice, this will entail a significant degree of virtual healthcare provision, which demands the exchange and accessibility of data. Collecting data in a standardised manner within the ERN framework will allow it to become syntactically and semantically interoperable, which increases the power of that data in several ways.

By the time ERNs were officially launched, certain sections of the rare disease field were already benefitting from the expertise of data management and IT fields: RD-ACTION therefore sought to introduce the full cohort of ERN coordinators to the ‘state of the art’ in data standardisation and data interoperability, to propose and debate on the application of several key resources/approaches which could be deployed in the ERN environment.

To this end, RD-ACTION co-organised a major workshop with DG Sante in Brussels on 25th and 26th April 2017. The workshop, entitled ‘Using Standards and Embedding Good Practices to Enable Interoperable Data-Sharing in ERNs’ united 68 participants from diverse fields.

Results: Four key resources/approaches were highlighted, as a starting point towards optimising the value of data in ERNs. The discussions evolved into a set of ‘Recommended Practices for Data Standardisation in the Context of the Operation of ERNs’. This poster illustrates these recommendations concerning how ERNs might embed these tools in their data-related activities. The recommendations deal with four essential -and interconnected- topics:

Coding Rare Diseases within ERNs

Capturing Phenotypic Information in ERNs

Demographic Data and Patient Pseudonymisation

Incorporating FAIR data principles to ERNs


P28 RD-Connect - an FP7 Success Story

Badowska D1, Wood L1, Thompson R1, Dawson J1, Lochmüller H2,3, Gut I3,4 and the RD-Connect Consortium

1John Walton Muscular Dystrophy Research Centre, Institute of Genetic Medicine, MRC Centre for Neuromuscular Diseases, Newcastle University, UK, 2Department of Neuropediatrics and Muscle Disorders, Medical Center–University of Freiburg, Faculty of Medicine, Freiburg, Germany 3Centro Nacional de Análisis Genómico (CNAG-CRG), Center for Genomic Regulation, Barcelona Institute of Science and Technology (BIST), Barcelona, Spain, 4Universitat Pompeu Fabra (UPF), Barcelona, Spain

Presenting author: Dorota Badowska Contact: [email protected]

Although individually uncommon, rare diseases (RDs) collectively affect 6-8% of the population. To address the unmet need of the rare disease community, in 2012, European Commission launched RD-Connect. In six years, the project has developed an infrastructure that integrates different types of data needed in rare disease research and facilitates their analysis and sharing across institutions and countries. The RD-Connect platform consists of three systems: Genome-Phenome Analysis Platform, Registry & Biobank Finder and Sample Catalogue, which already contain thousands of datasets including omics, phenotypes, biosamples and information on databases. The project has also developed several bioinformatic tools and data linage solutions as well as guidelines for ethical and patient-centred data sharing. The project helped move the field forward by advancing omics research and data sharing at their core in line with the goals of IRDiRC (International Rare Disease Research Consortium) and remains embedded in European and international collaborative efforts. Other important achievements include contribution to the discovery of over 100 rare disease genes, adaptation to the changing environment (European Reference Networks and General Data Protection Regulation) and creating a unique RD community that integrates scientists from different fields, clinicians and patients, who otherwise would have limited opportunities to work together. The success of RD-Connect makes it the flagship project of the European Commission.


31

P29 RD-Connect: data sharing and analysis for rare disease research within the integrated platform and through GA4GH Beacon and Matchmaker Exchange

Dawson J10, Beltran S1,2, Piscia D1,2, Laurie S1,2, Protasio J1,2, Papakonstantinou A1,2, Martinez I1,2, Valencia A3,14, Capella S3,14, Fernández JM3,14, Thompson M6, Kaliyaperumal R6, Lair S7, Sernadela P8, Girdea M9, Brudno M9, Blavier A7, Thompson R10, Lochmüller H10, Bellgard M11, Spalding D12, Roos M6, 't Hoen PAC6, Salgado D4,5, Béroud C4,5,13, Gut I1,2 and the RD-Connect Consortium

1Centro Nacional de Análisis Genómico (CNAG-CRG), Center for Genomic Regulation, Barcelona Institute of Science and Technology (BIST), Barcelona, Spain, 2Universitat Pompeu Fabra (UPF), Barcelona, Spain, 3Barcelona Supercomputing Center - Centro Nacional de Supercomputación (BSC-CNS), Barcelona, Spain, 4Aix-Marseille Université, Marseille, France, 5Inserm, UMR_S 910, Marseille, France, 6Department of Human Genetics, Leiden University Medical Center, Leiden, The Netherlands, 7Interactive Biosoftware, Rouen, France, 8DETI/IEETA, University of Aveiro, Portugal, 9Centre for Computational Medicine, Hospital for Sick Children and University of Toronto, Toronto, ON, Canada, 10Institute of Genetic Medicine, MRC Centre for Neuromuscular Diseases, Newcastle University, UK, 11Centre for Comparative Genomics, Murdoch University, Perth, Western Australia, 12European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Wellcome Trust Genome Campus, Cambridge, United Kingdom, 13APHM, Hôpital TIMONE Enfants, Laboratoire de Génétique Moléculaire, Marseille, France, 14Instituto Nacional de Bioinformática (INB), Spain

Presenting author: John Dawson Contact: [email protected]

The RD-Connect Genome-Phenome Analysis Platform for rare disease research brings together multiple omics data types (genomics, proteomics, transcriptomics) with biosample and clinical information at individual patient, family or whole-cohort level. It provides both a centralized data repository and a user-friendly online analysis system.

Whole-genome, exome and gene panel datasets are submitted by the end-user and processed by RD-Connect's standardised analysis and annotation pipeline to make data from different sequencing providers comparable. Clinical information is recorded in PhenoTips, simplifying clinical data entry using the Human Phenotype Ontology. Results are made available to the submitter and other authorised users through the highly configurable analysis interface (platform.rd-connect.eu) which enables filtering and prioritization of variants using common genomic location, effect, pathogenicity and population frequency annotations, enabling users to do their primary genomic analysis of their own patients online and compare with other submitted cohorts. Raw data is deposited at the European Genome-phenome Archive (EGA) for long-term storage.

The platform enables data sharing at various levels. At the most basic (“does this variant exist in this cohort?”) is the Global Alliance Beacon (www.beacon-network.org). At a more advanced level – finding patients in different databases with matching phenotype and candidate variant in the same gene – it is further developing Matchmaker Exchange (www.matchmakerexchange.org), allowing users of different systems to exchange information to find confirmatory cases. Finally, since patients have been consented for data sharing, authorized users can access datasets from other centres for further study.

The platform is open to any rare disease and already includes thousands of datasets from partner projects such as NeurOmics (www.rd-neuromics.eu) and BBMRI-LPC (www.bbmri-lpc.org). In 2018 it became the primary data sharing and analysis platform for the new Solve-RD project, which will bring in 19,000 unsolved cases from European Reference Networks over 5 years. RD-Connect is free and open for contributions from individual research groups and other projects: contact [email protected].

abstracts - rd-connect.eu · swertz ma1, monaco l2 1university medical center groningen, university...

Documents