MiMSI - a deep multiple instance learning framework improves microsatellite instability detection from tumor next-generation sequencing
Ziegler, J.; Hechtman, J. F.; Ptashkin, R.; Jayakumaran, G.; Middha, S.; Chavan, S. S.; Vanderbilt, C.; DeLair, D.; Casanova, J.; Shia, J.; DeGroat, N.; Benayed, R.; Ladanyi, M.; Berger, M. F.; Fuchs, T. J.; Zehir, A.
Show abstract
Microsatellite instability (MSI) is a critical phenotype of cancer genomes and an FDA-recognized biomarker that can guide treatment with immune checkpoint inhibitors. Recent work has demonstrated that next-generation sequencing data can be used to identify samples with MSI-high phenotype. However, low tumor purity, as frequently observed in routine clinical samples, poses a challenge to the sensitivity of existing algorithms. To overcome this critical issue, we developed MiMSI, an MSI classifier based on deep neural networks and trained using a dataset that included low tumor purity MSI cases in a multiple instance learning framework. On a challenging yet representative set of cases, MiMSI showed higher sensitivity (0.940) and auROC (0.988) than MSISensor(sensitivity: 0.57; auROC: 0.911), an open-source software previously validated for clinical use at our institution using MSK-IMPACT large panel targeted NGS data.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Clinical validation of Whole Genome Sequencing for routine cancer diagnostics 95%
- PanelCAT: an Open-Source Comparative Analysis Tool for Next-Generation Sequencing Panel Target Regions 94%
- Evaluating discordant somatic calls across mutation discovery approaches to minimize false negative drug-resistant findings 92%
Similar papers in this journal
Similar papers in this journal
- Tumor break load quantitates structural variant-associated genomic instability with biological and clinical relevance across cancers 93%
- EXaCT-2: An augmented and customizable oncology-focused whole exome sequencing platform 92%
- A Deep Learning Model for Molecular Label Transfer that Enables Cancer Cell Identification from Histopathology Images 92%
Similar papers in this journal
- Identification of single nucleotide variants using position-specific error estimation in deep sequencing data 93%
- Mutational profiling of micro-dissected pre-malignant lesions from archived specimens 93%
- Accuracy and Reproducibility of Somatic Point Mutation Calling in Clinical-Type Targeted Sequencing Data 92%
Similar papers in this journal
- Profiling diverse sequence tandem repeats in colorectal cancer reveals co-occurrence of microsatellite and chromosomal instability involving Chromosome 8 94%
- OncoGEMINI: Software for Investigating Tumor Variants From Multiple Biopsies With Integrated Cancer Annotations 94%
- Evaluating the transcriptional fidelity of cancer models 92%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.