Department of Radiation and Medical Oncology, Zhongnan Hospital of Wuhan University, 169 Donghu Road, Wuhan, 430071, Hubei, China.
Department of Biological Repositories, Zhongnan Hospital of Wuhan University, 169 Donghu Road, Wuhan, 430071, Hubei, China.
BMC Cancer. 2022 Dec 1;22(1):1243. doi: 10.1186/s12885-022-10339-3.
Radiotherapy has been widely used to treat various cancers, but its efficacy depends on the individual involved. Traditional gene-based machine-learning models have been widely used to predict radiosensitivity. However, there is still a lack of emerging powerful models, artificial neural networks (ANN), in the practice of gene-based radiosensitivity prediction. In addition, ANN may overfit and learn biologically irrelevant features.
We developed a novel ANN with Selective Connection based on Gene Patterns (namely ANN-SCGP) to predict radiosensitivity and radiocurability. We creatively used gene patterns (gene similarity or gene interaction information) to control the "on-off" of the first layer of weights, enabling the low-dimensional features to learn the gene pattern information. ANN-SCGP was trained and tested in 82 cell lines and 1,101 patients from the 11 pan-cancer cohorts.
For survival fraction at 2 Gy, the root mean squared errors (RMSE) of prediction in ANN-SCGP was the smallest among all algorithms (mean RMSE: 0.1587-0.1654). For radiocurability, ANN-SCGP achieved the first and second largest C-index in the 12/20 and 4/20 tests, respectively. The low dimensional output of ANN-SCGP reproduced the patterns of gene similarity. Moreover, the pan-cancer analysis indicated that immune signals and DNA damage responses were associated with radiocurability.
As a model including gene pattern information, ANN-SCGP had superior prediction abilities than traditional models. Our work provided novel insights into radiosensitivity and radiocurability.
放射治疗已被广泛用于治疗各种癌症，但疗效取决于个体。传统的基于基因的机器学习模型已被广泛用于预测放射敏感性。然而，在基于基因的放射敏感性预测实践中，仍然缺乏新兴的强大模型，即人工神经网络（ANN）。此外，ANN 可能会过拟合并学习与生物学无关的特征。
我们开发了一种基于基因模式的具有选择性连接的新型 ANN（即 ANN-SCGP），用于预测放射敏感性和放射可治愈性。我们创造性地使用基因模式（基因相似性或基因相互作用信息）来控制第一层权重的“开/关”，使低维特征能够学习基因模式信息。ANN-SCGP 在 82 个细胞系和来自 11 个泛癌队列的 1101 名患者中进行了训练和测试。
对于 2 Gy 时的存活分数，ANN-SCGP 的预测均方根误差（RMSE）在所有算法中最小（平均 RMSE：0.1587-0.1654）。对于放射可治愈性，ANN-SCGP 在 12/20 和 4/20 测试中分别获得了最大和第二大 C 指数。ANN-SCGP 的低维输出再现了基因相似性的模式。此外，泛癌分析表明，免疫信号和 DNA 损伤反应与放射可治愈性相关。
作为一种包含基因模式信息的模型，ANN-SCGP 具有比传统模型更好的预测能力。我们的工作为放射敏感性和放射可治愈性提供了新的见解。
