A research team led by Dr. Altuna Akalin of the Max Delbrück Center in Berlin has developed an innovative tool called Flexynesis that helps cancer patients and their doctors find the optimal therapy. The tool uses deep neural networks to simultaneously evaluate multimodal data such as multi-omics data, processed texts and medical image data, for example from CT or MRI images. In this way, it enables more precise diagnoses, prognoses and tailor-made treatment strategies.

The number of new cancer therapies is growing steadily – almost fifty new treatment methods are added every year. However, this diversity makes it difficult to select the best therapy for patients with individual tumor characteristics. Flexynesis addresses this problem by integrating different data types and analyzing them simultaneously. For example, the tool can determine the exact origin of cancer, identify effective drugs and assess their impact on the chances of survival. It also supports the search for biomarkers for diagnosis and prognosis, for example in the case of metastases of unclear origin.

Flexynesis was developed by Akalin’s team, with Dr. Bora Uyar playing a major role as first and co-corresponding author of the publication. Unlike many existing deep learning tools, which are often inflexible or difficult to implement, Flexynesis is versatile and accessible through platforms such as PyPI, Guix, Docker, Bioconda, and Galaxy. This facilitates integration into existing work processes of clinics and research institutions.

Compared to other tools, Flexynesis stands out for its ease of use and flexibility. It makes it possible to answer various medical questions, such as the type of cancer, the effectiveness of drugs or the origin of tumors. In addition to Flexynesis, Akalin’s team had previously developed another AI tool called Onconaut, which is based on known biomarkers, clinical trials, and guidelines. Both tools can complement each other in a meaningful way.

One challenge remains the availability of multi-omics data in clinics, which in Germany have so far only been used in special programs such as the MASTER program for rare cancers. In the USA, on the other hand, such data have already been established in tumor conferences to support therapy decisions. Flexynesis does not require extensive AI knowledge, which makes it easier for doctors and clinical researchers to use. Online instructions also support the application.

The continuously updated tool promises to lower the barriers to integrating multimodal data into clinical practice and promote the development of personalized treatment strategies for cancer patients.

Code:

GitHub – BIMSBbioinfo/flexynesis: A deep-learning based multi-modal data integration suite that aims to achieve synesis in a flexible manner

Original Paper:

Bora Uyar, et al. (2025): “Flexynesis: A deep learning toolkit for bulk multi-omics data integration for precision oncology and beyond”. Nature Communications, DOI: 10.1038/s41467-025-63688-5

Editor: X-Press Journalistenbüro GbR

Gender Notice. The personal designations used in this text always refer equally to female, male and diverse persons. Double/triple naming and gendered designations are used for better readability. ected.