The analysis of high-resolution sequencing and multi-omics data forms the basis of modern precision oncology – but it is only through complex bioinformatic processes that raw data is transformed into clinically actionable knowledge. This is precisely where the Computational Oncology (CO) research group comes in: it develops and operates the bioinformatic infrastructure that makes other key activities possible:

• the molecular tumour boards (MTBs) and
• the large-scale precision oncology programmes of the National Centre for Tumour Diseases (NCT) and the German Cancer Research Centre (DKFZ)
• molecularly stratified clinical trials

CO is part of the Molecular Precision Oncology Programme (led by Stefan Fröhling) and is co-led by Daniel Hübschmann and Marc Zapatka.

PD Dr Dr Daniel Hübschmann is a physician and bioinformatician and heads the Innovation and Service Unit for Bioinformatics and Precision Medicine at the DKFZ; together with Marc Zapatka, he leads the Computational Oncology research group at the NCT/DKFZ in Heidelberg, as well as the former early-career research group ‘Pattern Recognition & Digital Medicine’ at the Heidelberg Institute for Stem Cell Technology and Experimental Medicine (HI-STEM). His research focuses on pattern recognition in high-dimensional genomic data, the integration of clinical, multi-omics and single-cell data for precision oncology, the quantitative analysis of cellular interactions, and digital decision-support and visualisation systems for molecular tumour boards.

Dr Marc Zapatka is a bioinformatician and, together with Daniel Hübschmann, heads the Computational Oncology research group. His work focuses on computational cancer genomics: the analysis of next-generation sequencing data, the identification of driver mutations and structural variants, and the development of robust, reproducible analysis pipelines for large cohorts. He is involved in numerous international collaborative projects, including those on the landscape of viral associations in human cancers and on paediatric tumours.

a) Bioinformatics for molecular tumour boards, precision oncology programmes and clinical trials

CO provides the bioinformatics analysis infrastructure for key precision oncology programmes of the DKTK/NCT:

• MASTER (Molecularly Aided Stratification for Tumour Eradication Research): a programme for patients with advanced rare tumours or unusually early-onset, otherwise incurable cancers. It combines genomic, transcriptomic and methylation analyses with rapid clinical feedback to the molecular tumour boards at the NCT sites in Heidelberg and Dresden, as well as other DKTK partner sites.
• CATCH (Comprehensive Assessment of Clinical Features and Biomarkers To Identify Patients with Advanced or Metastatic Breast Cancer for Marker-Driven Trials in Humans): creates genotype and biomarker profiles of metastases in advanced/metastatic breast cancer to identify therapy-relevant driver mutations to inform treatment decisions.
• COGNITION: (Comprehensive assessment of clinical features, genomics and further molecular markers to identify patients with early breast cancer for enrolment on marker-driven trials) applies the same approach to early-stage breast cancer in patients with an inadequate response to neoadjuvant therapy and forms the molecular basis for the subsequent intervention study COGNITION-GUIDE.

The analyses provided are also used for clinical trials, including those from the NCT’s OCT2 programme:

• COGNITION-GUIDE 
  Trial Details: https://clinicaltrials.gov/study/NCT05332561
• CRAFT
  Trial Details: https://clinicaltrials.gov/study/NCT04551521
• RATIONALE 
  Trial Details: https://clinicaltrials.gov/study/NCT06855134
• TOP-ART 
  Trial Details: https://clinicaltrials.gov/study/NCT03127215

The bioinformatics workflow encompasses the fundamental steps of NGS data analysis – alignment and quality control, somatic variant calling, and the pre-filtering of clinically relevant and/or therapeutically targetable mutations. Based on the somatic variants of each tumour, a comprehensive annotation is carried out against databases of drug targets, known driver mutations and germline predisposition variants. The interpretation prioritises potentially therapeutically relevant variants and links them to systemic cancer therapies; drawing on oncological expertise, the level of evidence from the literature and clinical trials is determined for each candidate mutation.

To accelerate and standardise this process, CO, in collaboration with the Secondary Use of Data in Oncology Group, has developed and operates the Knowledge Connector – a digital decision-support system that presents molecular findings in a structured format, links them to up-to-date specialist knowledge, and thus enables molecular tumour boards to carry out evidence-based, efficient assessments. The system is underpinned by an innovative knowledge base, , known as the BoCKbase. This is based on the ‘Block of Clinical Knowledge’ (BoCK) concept, an abstraction of the biomarker concept that enables the linking of evidence building blocks. The system has been in use since 2022 at the NCT sites in Heidelberg, Dresden, Berlin and WERA (Augsburg) and has already processed over 5,000 patient cases.

b) Cohort analyses and scientific projects

In addition to MTB analytics, CO is responsible for NGS analyses covering a broad spectrum of tumour types and data types. The aim is to provide standardised data processing and analysis workflows that reflect the current state of the art. CO is currently developing and operating workflows for the centralised, standardised processing of:

• Alignment of whole-genome, whole-genome bisulphite, exome, ChIP and RNA sequencing data (as well as variants of these technologies)
• somatic variant calling for single nucleotide variants (SNVs), insertions and deletions (indels), structural variants (SVs) and copy number alterations (CNAs)
• methylation analyses

c) Method and tool development

In addition, CO develops resources for exploring high-dimensional datasets, self-learning cohort analysis tools and machine learning methods for multi-omics and single-cell data. The group is committed to responsible and reproducible bioinformatics practice and, together with the Clinical Bioinformatics team, organises training courses on clinical bioinformatics analysis.

CO develops and maintains its code base in close collaboration with the Omics IT and Data Management Core Facility (ODCF, Head: Ivo Buchhalter) and the Department of Applied Bioinformatics (ABI, Head: Benedikt Brors). As part of the molecular tumour boards, CO works closely with the clinical departments of Translational Medical Oncology (TMO, Head: Stefan Fröhling) and Gynaecological Oncology / Breast Cancer (Head: Andreas Schneeweiss).

Listed in alphabetical order

Lab Management
Dr Katja Weichsel

Researchers
Dr German Demidov
Fangyoumin Feng
Dr Martina Fröhlich
Dr Mario Hlevnjak
Dr Barbara Hutter
Dr Jennifer Hüllein
Carola Kensche
Polina Kozyulina
Dr Kübra Narci
Dr Malgorzata Oles
Dr Rajesh Pal
Dr Nagarajan Paramasivam
Dr Lorenzo Rigano
Dr Marc Rübsam
Dr David Schlütermann
Dr Anamika Thalor
Dr Ciğdem Hazal Timucin
Dr Shen Zhong

Quality Management Officers
Saskia Egarter
Carola Leitner

PhD students
Anastasia Sedlmeier
Sunidhi Sunidhi

Associate members, students
Waqar Hussain
Celine Krogmann
Yejin Kwak
Marco Rheinnecker
Giulia Roeth
Angelo Jovin Yamachui Sitcheu 

Alumni
Dr Jude al-Sabah
Dr Carolin Andresen
Dr Zuguang Gu
Dr Lea Jopp-Saile
Dr Sebastian Pirmann
Dr Sebastian Uhrig