Transformer for Gene Expression Modeling (T-GEM): An Interpretable Deep Learning Model for Gene Expression-Based Phenotype Predictions.
Transformer for Gene Expression Modeling (T-GEM): An Interpretable Deep Learning Model for Gene Expression-Based Phenotype Predictions.
复制标题
作者：
癌症是全球第二大死亡原因。预测表型和理解定义表型的标记是重要的任务。我们提出了一种名为T-GEM的可解释深度学习模型，可以预测癌症相关表型预测并揭示表型相关的生物学功能和标记基因。我们证明了T-GEM使用TGCA数据预测癌症类型和使用scRNA-seq数据识别免疫细胞类型的能力。提供的代码和详细的文件，以方便在其他研究中的模型容易实施。深度学习已应用于精确肿瘤学，以解决各种基于基因表达的表型预测。然而，基因表达数据的独特特征挑战了流行的深度学习（DL）模型（如卷积神经网络（CNN））的计算机视觉灵感设计，并要求开发为转录组学研究量身定制的可解释的DL模型。为了解决当前在开发用于建模基因表达数据的可解释的DL模型方面的挑战，我们提出了一种新的可解释的深度学习架构，称为T-GEM，或用于基因表达建模的Transformer。我们提供了详细的T-GEM模型用于建模基因-基因相互作用，并证明了其在基于基因表达的癌症相关表型预测中的实用性，包括癌症类型预测和免疫细胞类型分类。我们仔细分析了T-GEM的学习机制，发现第一层有更广泛的关注，而更高的层更关注表型相关的基因。我们还表明，T-GEM的自我注意力可以捕获与预测的表型相关的重要生物学功能。我们进一步设计了一种方法来提取T-GEM通过利用自我注意力权重的属性进行分类来学习的调控网络，并表明网络中心基因可能是预测表型的标记。
Cancer is the second leading cause of death worldwide. Predicting phenotype and understanding makers that define the phenotype are important tasks. We propose an interpretable deep learning model called T-GEM that can predict cancer-related phenotype prediction and reveal phenotype-related biological functions and marker genes. We demonstrated the capability of T-GEM on cancer type prediction using TGCA data and immune cell type identification using scRNA-seq data. The code and detailed documents are provided to facilitate easy implementation of the model in other studies. Deep learning has been applied in precision oncology to address a variety of gene expression-based phenotype predictions. However, gene expression data’s unique characteristics challenge the computer vision-inspired design of popular Deep Learning (DL) models such as Convolutional Neural Network (CNN) and ask for the need to develop interpretable DL models tailored for transcriptomics study. To address the current challenges in developing an interpretable DL model for modeling gene expression data, we propose a novel interpretable deep learning architecture called T-GEM, or Transformer for Gene Expression Modeling. We provided the detailed T-GEM model for modeling gene–gene interactions and demonstrated its utility for gene expression-based predictions of cancer-related phenotypes, including cancer type prediction and immune cell type classification. We carefully analyzed the learning mechanism of T-GEM and showed that the first layer has broader attention while higher layers focus more on phenotype-related genes. We also showed that T-GEM’s self-attention could capture important biological functions associated with the predicted phenotypes. We further devised a method to extract the regulatory network that T-GEM learns by exploiting the attributions of self-attention weights for classifications and showed that the network hub genes were likely markers for the predicted phenotypes.
登录
查看更多内容
影响因子：
14.9
作者：
Colaprico A;Silva TC;Olsen C;Garofano L;Cava C;Garolini D;Sabedot TS;Malta TM;Pagnotta SM;Castiglioni I;Ceccarelli M;Bontempi G;Noushmehr H
通讯作者：
Noushmehr H
影响因子：
5.6
作者：
Banegas-Luna AJ;Peña-García J;Iftene A;Guadagni F;Ferroni P;Scarpato N;Zanzotto FM;Bueno-Crespo A;Pérez-Sánchez H
通讯作者：
Pérez-Sánchez H
影响因子：
5.2
作者：
Haam S;Han JH;Lee HW;Koh YW
通讯作者：
Koh YW
影响因子：
2.7
作者：
Chiu, Yu-Chiao;Chen, Hung-I Harry;Chen, Yidong
通讯作者：
Chen, Yidong
影响因子：
64.8
作者：
Elmarakeby HA;Hwang J;Arafeh R;Crowdis J;Gang S;Liu D;AlDubayan SH;Salari K;Kregel S;Richter C;Arnoff TE;Park J;Hahn WC;Van Allen EM
通讯作者：
Van Allen EM