Objective

HPV-associated head and neck cancer is correlated with favorable prognosis; however, its underlying biology is not fully understood. We propose an explainable convolutional neural network (CNN) classifier, DeepClassPathway, that predicts HPV-status and allows patient-specific identification of molecular pathways driving classifier decisions.

Methods

The CNN was trained to classify HPV-status on transcriptome data from 264 (13% HPV-positive) and tested on 85 (25% HPV-positive) head and neck squamous carcinoma patients after transformation into 2D-treemaps representing molecular pathways. Grad-CAM saliency was used to quantify pathways contribution to individual CNN decisions. Model stability was assessed by shuffling pathways within 2D-images.

Results

The classification performance of the CNN-ensembles achieved ROC-AUC/PR-AUC of 0.96/0.90 for all treemap variants. Quantification of the averaged pathway saliency heatmaps consistently identified KRAS, spermatogenesis, bile acid metabolism, and inflammation signaling pathways as the four most informative for classifying HPV-positive patients and MYC targets, epithelial–mesenchymal transition, and protein secretion pathways for HPV-negative patients.

Conclusion

We have developed and applied an explainable CNN classification approach to transcriptome data from an oncology cohort with typical sample size that allows classification while accounting for the importance of molecular pathways in individual-level decisions.

中文翻译：

---

DeepClassPathway：使用可解释的深度学习进行分子通路感知分类

客观的

HPV 相关的头颈癌与良好的预后相关；然而，其基础生物学尚未完全了解。我们提出了一种可解释的卷积神经网络 (CNN) 分类器 DeepClassPathway，它可以预测 HPV 状态并允许对驱动分类器决策的分子通路进行患者特异性识别。

方法

CNN 接受了训练，可以根据 264 名（13% HPV 阳性）的转录组数据对 HPV 状态进行分类，并在转化为代表分子通路的二维树图后对 85 名（25% HPV 阳性）头颈部鳞癌患者进行测试。Grad-CAM 显着性用于量化通路对单个 CNN 决策的贡献。通过 2D 图像内的改组路径评估模型稳定性。

结果

对于所有树图变体，CNN 集成的分类性能达到 0.96/0.90 的 ROC-AUC/PR-AUC。平均通路显着性热图的量化始终将 KRAS、精子发生、胆汁酸代谢和炎症信号通路确定为对 HPV 阳性患者和 MYC 靶点、上皮-间质转化和 HPV 阴性患者的蛋白质分泌通路进行分类的四种最有用信息.

结论

我们开发了一种可解释的 CNN 分类方法并将其应用于具有典型样本量的肿瘤队列的转录组数据，该方法允许分类，同时考虑分子途径在个体层面决策中的重要性。