Johannes Starlinger

I'm a computer science postdoc and MD doing research in biomedical informatics at the Department of Anesthesiology and Operative Intensive Care Medicine at Charité – Universitätsmedizin Berlin and the research group for Knowledge Management in Bioinformatics at the Department of Computer Science at Humboldt-Universität zu Berlin. After studying medicine at Medical University of Vienna and computer science at HU-Berlin, I joined the DFG-funded graduate program SOAMED in 2010 to research service-oriented architectures in a medical area of application, receiving my PhD in 2015. My current research focus is on knowledge mining and representation, similarity search, and predictive analytics over data relevant to the biomedical domain, including genomic and clinical data. I'm PI on the simpatix project, funded by a DFG "Temporary Position for Principal Investigators" (Eigene Stelle), technical coordinator in PREDICT, and member of the ASTRODEM project.

Email: starlinger (at) informatik.hu-berlin.de

since 01/2018

Health data researcher at Charité – Universitätsmedizin Berlin
Funded by a DFG-grant for "Temporary Positions for Principal Investigators" (Eigene Stelle)

Visiting researcher at Humboldt-Universität zu Berlin

05/2015 – 12/2017

Postdoctoral researcher at Humboldt-Universität zu Berlin
Funded by a DFG-grant for "Temporary Positions for Principal Investigators" (Eigene Stelle) from October 2016

08/2010 – 05/2015

PhD Student in the DFG-funded graduate school SOAMED (Dr.-Ing.)
Humboldt-Universität zu Berlin
Department of Computer Science
Knowledge Management in Bioinformatics

09/2012 – 11/2012

Visiting scholar at the Department of Computer and Information Science at the University of Pennsylvania with Prof Susan B. Davidson
Funded by a DAAD short term grant for PhD students

10/2005 – 08/2010

Diploma in Computer Science (Dipl.-Inf.)
Humboldt-Universität zu Berlin

2010 Diploma Thesis "Integration of clinical record data with special consideration of diagnosis classification"

2009 Study Project "Integration and visualization of heterogeneous annotation results of the BioCreAtIvE Meta-Server"

11/2008 – 07/2010 Student assistant at the research group for Knowledge Management in Bioinformatics at Humboldt University

01/2007 – 01/2009 Student assistant at the Institute of Cultural Sciences at Humboldt University

10/1998 – 05/2005

Medical studies (Dr. med. univ.)
Medical University of Vienna

2000 – 2004 Tutor and Research Assistant at the Institute of Anatomy, University of Vienna / Medical University of Vienna with Prof. Helmut Gruber

08/2003 – 09/2003 Research visit at the University of Cincinnati College of Medicine Department of Surgery in the Laboratory of Epithelial Pathobiology with Prof. Jeffrey B. Matthews and Roger T. Worrell

• Similiarity measures and similiarty search over process structured data such as scientific workflows, business processes, or patient histories
• Text mining and data analytics for medical and especially clinical documents
• Dataspaces, Pay-as-you-go Data Integration

Ongoing

simpatix

In the simpatix project, we investigate similarity search over electronic health records. These records consist of mostly unstructured or semi-structured data, such as clinical notes from examinations and treatments, tabularized data from quantitative tests (such as blood screenings), or discharge summaries, but encode an implicit process describing the individual patient’s disease history. In simpatix, we extract this process from EHRs, together with rich annotations of clinically relevant entities (e.g., diagnoses, treatments, or procedures), and investigate similarity measures for such process-structured case representations to compare and find similar cases. In the end, we want to deploy these measures in similarity search over large collections of patient cases to enable use cases such as clinical decision support.

PREDICT

Our aim in PREDICT is to develop a software system that enables clinicians to use the large body of existing data on the relationships between genetic/epigenetic alterations on the one hand, and treatment options in cancer and their success on the other hand. We are developing a semantically integrated cancer-type specific knowledge base drawing its data from a variety of data sources, including scientific publications, genotype/phenotype databases, drug effectiveness screens, clinical trials, large-scale association studies etc., using advanced and innovative algorithms for knowledge extraction, semantic data integration, and biomedical text mining.

ASTRODEM

Indentifying predictive markers for the onset of dementia and creating predictive models which will help general practitioners (GPs) identify patients at high risk of dementia is the aim of the ASTRODEM project at Brighton and Sussex Medical School, using 96,000 anonymised GP patient records from the Clinical Practice Research Datalink (CPRD). Collaborating with Dr Elizabeth Ford and her team as a visiting researcher, my special interest is making use of longitudinal information include in the patient records.

Finished

FlowAlike

In the FlowAlike project we collected a corpus of similarity ratings for scientific workflows from a dataset of 1483 Taverna workflows and a second dataset of 139 Galaxy workflows, to be used for the evaluation of algorithmic similarity measures. Around 2400 ratings were manually assigned by scientific workflow experts, and are available for download to test and compare your own algorithms!

GeneView

GeneView is a web-based tool for searching and visualizing a deeply annotated copy of PubMed and the open access subset of PubMed Central. While I was majorly resonsible for creating the web-interface during my time as a student assistant in the research group for Knowledge Management in Bioinformatics at HU-Berlin, Philippe Thomas is the tool's main developer and researcher behind it. Check it out!

PiPa

PiPa is a Java-based tool to setup and maintain a comprehensive database of information on protein-protein interactions and biological pathways, integrated from a range of multiple public protein information databases. The data is stored in a local MySQL database to provide programmatic access through standard SQL query APIs. The tool is still available for download.

BC-VisCon

The BC-VisCon tool was created as part of my study project "Integration and visualization of heterogeneous annotation results of the BioCreAtIvE Meta-Server".
Goal of the BC-VisCon website is to provide an intuitive and flexible tool for viewing and analyzing aggregations and comparisons of the annotations made by the gene mention tagging servers connected to the BioCreAtIve-MetaServer.

Summer term 2017

Lecture Introduction to Bioinformatics

Winter term 2016/17

Guest lecture on Process Similarity Search as part of Ulf Leser's lecture Implemantation of Database Systems

Summer term 2016

Lecture unit on Protein Function and Structure Prediction as part of Ulf Leser's lecture Introduction to Bioinformatics

Summer term 2015

Seminar Informatik in der Medizin

Exercises following the course Grundlagen der Bioinformatik
Together with Ulf Leser

Summer term 2013

Seminar Similarity Search
Together with Ulf Leser, Sebastian Wandelt, Andre Koschmieder, and Astrid Rheinländer

Summer term 2012

Seminar Large Scale Data Analysis
Together with Ulf Leser, Mark Bux, and Astrid Rheinländer

Lecture unit on "Creating a basic web2.0-application using XHTML, CSS, JavaScript and PHP"
Part of the Ringvorlesung zu Semesterprojekten
Material: Slides and Code

Scientific Service

• External Reviewer for BMC Bioinformatics, ICDE, EDBT, SMBM, LWDA
• Reviewer for VLDBJ, TKDE, Journal of Biomedical Semantics, Science China, DESRIST

Workshops

2017

Co-organizer of the workshop on Extracting evidence from clinical free text: opportunities and challenges at the Informatics for Health Conference, Manchester, UK