Related Experiment Video
Updated: Jul 4, 2025
DNA-barcode-based Multiplex Immunofluorescence Imaging to Analyze FFPE Specimens from Genetically Reprogrammed Murine Melanoma
Published on: June 6, 2025
DEEPOMICS FFPE, a deep neural network model, identifies DNA sequencing artifacts from formalin fixed paraffin
Dong-Hyuk Heo1, Inyoung Kim1, Heejae Seo1
1Theragen Bio Co., Ltd., Seongnam, Gyeonggi-do, 13488, Republic of Korea.
Abstract:
Formalin-fixed, paraffin-embedded (FFPE) tissue specimens are routinely used in pathological diagnosis, but their large number of artifactual mutations complicate the evaluation of companion diagnostics and analysis of next-generation sequencing data. Identification of variants with low allele frequencies is challenging because existing FFPE filtering tools label all low-frequency variants as artifacts. To address this problem, we aimed to develop DEEPOMICS FFPE, an AI model that can classify a true variant from an artifact. Paired whole exome sequencing data from fresh frozen and FFPE samples from 24 tumors were obtained from public sources and used as training and validation sets at a ratio of 7:3. A deep neural network model with three hidden layers was trained with input features using outputs of the MuTect2 caller. Contributing features were identified using the SHapley Additive exPlanations algorithm and optimized based on training results. The performance of the final model (DEEPOMICS FFPE) was compared with those of existing models (MuTect filter, FFPolish, and SOBDetector) by using well-defined test datasets. We found 41 discriminating properties for FFPE artifacts. Optimization of property quantification improved the model performance. DEEPOMICS FFPE removed 99.6% of artifacts while maintaining 87.1% of true variants, with an F1-score of 88.3 in the entire dataset not used for training, which is significantly higher than those of existing tools. Its performance was maintained even for low-allele-fraction variants with a specificity of 0.995, suggesting that it can be used to identify subclonal variants. Different from existing methods, DEEPOMICS FFPE identified most of the sequencing artifacts in the FFPE samples while retaining more of true variants, including those of low allele frequencies. The newly developed tool DEEPOMICS FFPE may be useful in designing capture panels for personalized circulating tumor DNA assay and identifying candidate neoepitopes for personalized vaccine design. DEEPOMICS FFPE is freely available on the web ( http://deepomics.co.kr/ffpe ) for research.
Insights
A new AI model, DEEPOMICS FFPE, accurately distinguishes true genetic variants from artifacts in formalin-fixed, paraffin-embedded (FFPE) tissues. This tool improves next-generation sequencing analysis by preserving low-frequency variants crucial for diagnostics and research.
Area of Science:
- Genomics
- Bioinformatics
- Computational Biology
Background:
- Formalin-fixed, paraffin-embedded (FFPE) tissues are vital for pathology but introduce artifactual mutations.
- Existing tools struggle to differentiate true variants from FFPE artifacts, especially at low allele frequencies.
- Accurate variant identification is critical for companion diagnostics and next-generation sequencing (NGS) data analysis.
Purpose of the Study:
- To develop DEEPOMICS FFPE, an artificial intelligence (AI) model for classifying true variants from FFPE artifacts.
- To improve the accuracy of variant detection in FFPE samples for NGS analysis.
- To provide a tool that can identify subclonal variants and aid in personalized medicine applications.
Main Methods:
- Utilized paired whole exome sequencing data from fresh frozen and FFPE tumor samples for training and validation.
- Developed a deep neural network model incorporating features from MuTect2 caller outputs.
- Employed SHapley Additive exPlanations (SHAP) for feature identification and optimization.
- Compared DEEPOMICS FFPE performance against existing tools like MuTect filter, FFPolish, and SOBDetector.
Main Results:
- DEEPOMICS FFPE achieved an F1-score of 88.3, removing 99.6% of artifacts while retaining 87.1% of true variants.
- Demonstrated high specificity (0.995) for low-allele-fraction variants, enabling subclonal variant identification.
- Outperformed existing FFPE artifact filtering tools in accuracy and variant retention.
Conclusions:
- DEEPOMICS FFPE effectively identifies sequencing artifacts in FFPE samples while preserving true variants, including low-frequency ones.
- The tool shows significant potential for designing capture panels for circulating tumor DNA assays and identifying neoepitopes for personalized vaccines.
- DEEPOMICS FFPE is freely available online, facilitating research in personalized oncology.
