Forome Anfisa – an open source variant interpretation tool

Whole exome and whole genome sequencing are being rapidly adopted in the healthcare industry, making way into the routine clinical practice. Most variant interpretation tools are built to work with domain-based clinical guidelines, approved by ACMG and other responsible bodies, focusing on minimizing the number of variants for manual review by clinical staff. Forome Anfisa is a collaborative variant annotation, interpretation, and curation tool, an organic part of the integrated clinical research program, developed by the Forome Platform team. Initially built as part of Brigham Genomics Medicine Program and is used for a hearing loss project (SEQaBOO), Forome Anfisa is now re-architected for whole exome and genome cases, enabling smooth work with a huge volume of data and giving way for clinicians to cope with millions of genetic variants in a meaningful way. Introducing the first fully open source variant management toolset aimed at both clinical and research communities, we provide a way to seamlessly transform research workflows into clinical guidelines, thus speeding up the adoption of WGS/WES into the clinical practice. It is offering collaboration by design, allowing interaction of users in different roles and sharing of findings in a particular patient case or even a variant. Forome Anfisa is a highly customizable suite of software for downstream genetic analysis, clinical variant interpretation, curation, and collaboration. It supports 3 real-life scenarios for effective WES/WGS and panel of genes variants analysis: the traditional clinical workflow for variant curation based on predefined guidelines; a workflow for design and development new guidelines for variant interpretation; a collaboration in variant interpretations. Anfisa is developed under the Apache 2.0 license and is available on GitHub in the Forome Association repository: https://github.com/ForomePlatform/anfisa . Anfisa is a modular system with three main components and a number of support modules. Main components are: annotation pipeline, backend database, frontend user interface. Annotation pipeline starts with Ensembl VEP [1] and then adds annotations based on functional analysis, population genetics, clinical knowledge, epigenetics, etc. This is done by traversing databases such as gnomAD, ClinVar [2], HGMD [3], including results from spliceAI [4] and other sources. The backend stores the data in Druid Open Source OLAP [5] and metadata, such as user environment, curation notes and preferences are stored in MongoDB. The frontend is implemented using Vue JavaScript Framework [6] and Bootstrap toolkit [7]. Annotation Pipeline and Backend provide public REST API and technically can be used in standalone mode, integrated with other genomics tools and EMRs. Anfisa is designed to forge collaboration between people with different goals and skills, and with different organizational roles, from treating physicians to clinical geneticists to researchers and bioinformaticians. The system is designed to efficiently operate with small and large genomic datasets. It is transparent for users.  A patient case consisting of the panel of genes loads several thousands variants directly into the main UI and offers filtering capabilities to quickly narrow the list down to a few dozens. It is then a workable amount of data to review manually. A case with WES/WGS data loads into the advanced filtering tool which would help users creating a custom workspace. The workspace is a combination of the various available filters and more complex rules (clinical guidelines), used by the user to reach a reasonable and meaningful amount of the variants to work in the manual mode. When the workspace is defined and applied to the case data, a list of variants is loaded into the main UI. We have implemented two distinct scenarios to address the variety of users goals of the complexity of clinical research tasks:  the clinical use, where the users operate with predefined clinical rules (e.g. ACMG guidelines) and thus, assuring the standardized protocols in the clinical practice are being followed; the research scenario where the new rules are being evaluated and tested and eventually being promoted for use in the clinical scenario. This approach implements our collaboration concept to bring the research findings quicker to the clinical practice. It is specifically valuable for WGS/WES diagnostics where the guidelines are still evolving.       Interpretation mode Research Mode The clinical scenario implies working within the set of pre-defined clinical guidelines. It helps to focus on efficiency and collaboration. It allows to tag, comment and eventually to report on the candidate variants. This scenario is being used by SEQaBOO Project [8]. The researcher scenario allows a clinical geneticist or a researcher to apply various flexible criteria and create workspaces that can be shared with other collaborating researches. Collaborating researches can tag specific variants in the workspace that were shared with them and put textual notes explaining their reasoning.  A special workflow is designed for developing new clinical guidelines. This workflow allows the user to build a decision tree for variant classification.  Advanced workflows are being adopted by Brigham Genomics Medicine Project [9] References  McLaren W, et al. The Ensembl Variant Effect Predictor. Genome Biol. 2016;17:122. doi: 10.1186/s13059-016-0974-4.  Landrum M.J., Lee J.M., Benson M. et al. (2018) ClinVar: improving access to variant interpretations and supporting evidence. Nucleic Acids Res., 46, D1062–D1067. Stenson P. D., Ball E. V., Mort M., Phillips A. D., Shaw K., Cooper D. N. (2012). The Human Gene Mutation Database (HGMD) and its exploitation in the fields of personalized genomics and molecular evolution. Curr. Protoc. Bioinformatics 39 1.13.1–1.13.20.  K. Jaganathan, S. Kyriazopoulou Panagiotopoulou, J.F. McRae, S.F. Darbandi, D. Knowles, Y.I. Li, J.A. Kosmicki, J. Arbelaez, W. Cui, G.B. Schwartz, et al. Predicting Splicing from Primary Sequence with Deep Learning. Cell, 176 (2019), pp. 535-548.e24 Apache Druid is a high performance real-time analytics database http://druid.io/docs/latest/design/   Vue - The Progressive JavaScript Framework  https://vuejs.org/   Bootstrap - an open source toolkit for developing with HTML, CSS, and JS. https://getbootstrap.com/   Haghighi A., Krier J.B., Toth-Petroczy A., Cassa C.A., Frank N.Y., Carmichael N., Fieg E., Bjonnes A., Mohanty A., Briere L.C.et al.  An integrated clinical program and crowdsourcing strategy for genomic sequencing and Mendelian disease gene discovery. NPJ Genome Med.  2018  SEQaBOO - SEQuencing a Baby for an Optimal Outcome (http://seqaboo.bwh.harvard.edu/), under the NIH grant R01-DC015052-01