Title: Flatsomatic: A Method for Compression of Somatic Mutation Profiles in Cancer Show Chinese title

Title: Flatsomatic：一种用于癌症体细胞突变图谱压缩的方法

Abstract: In this study, we present Flatsomatic - a Variational Auto Encoder (VAE) optimized to compress somatic mutations that allow for unbiased data compression whilst maintaining the signal. We compared two different neural network architectures for the VAE: Multilayer Perceptron (MLP) and bidirectional LSTM. The somatic profiles we used to train our models consisted of 8,062 Pan-Cancer patients from The Cancer Genome Atlas and 989 cell lines from the COSMIC cell line project. The profiles for each patient were represented by the genomic loci where somatic mutations occurred and, to reduce sparsity, the locations with a frequency <5 were removed. We enhanced the VAE performance by changing its evidence lower bound, and devised an F1-score based loss showing that it helps the VAE learn better than with binary cross-entropy. We also employed beta-VAE to weight the variational regularisation term in the loss function and showed the best performance through a preliminary function to increase the weight of the regularisation term with each epoch. We assessed the reconstruction ability of the VAE using the micro F1-score metric and showed that our best performing model was a 2-layer deep MLP VAE. Our analysis also showed that the size of the latent space did not have a significant effect on the VAE learning ability. We compared the Flatsomatic embeddings created to a lower dimension version of the data from principal component analysis, showing superior performance of Flatsomatic, and performed K-means clustering on both datasets to draw comparisons to known cancer types of each profile. Finally, we present results that confirm that the Flatsomatic representations of 64 dimensions maintain the same predictive power as the original 8,298 dimensions vector, through prediction of drug response. Show Chinese abstract

Abstract: 在本研究中，我们提出了Flatsomatic - 一种优化的变分自编码器（VAE），用于压缩体细胞突变，允许无偏的数据压缩同时保持信号。 我们比较了两种不同的VAE神经网络架构：多层感知器（MLP）和双向LSTM。 我们用于训练模型的体细胞特征包括来自The Cancer Genome Atlas的8,062名泛癌患者和来自COSMIC细胞系项目的989个细胞系。 每个患者的特征由体细胞突变发生的基因组位置表示，并为了减少稀疏性，移除了频率<5的位置。 我们通过改变其证据下界来增强VAE性能，并设计了一个基于F1分数的损失函数，表明它有助于VAE比二元交叉熵更好地学习。 我们还使用了beta-VAE来加权损失函数中的变分正则化项，并通过一个初步函数在每个周期中增加正则化项的权重，展示了最佳性能。 我们使用微F1分数度量评估VAE的重建能力，并表明我们表现最好的模型是一个两层深度MLP VAE。 我们的分析还显示，潜在空间的大小对VAE的学习能力没有显著影响。 我们将Flatsomatic生成的嵌入与主成分分析的低维数据版本进行了比较，显示出Flatsomatic的优越性能，并对两个数据集进行了K-means聚类，以将每个特征与已知的癌症类型进行比较。 最后，我们展示了结果，证实了64维的Flatsomatic表示保持与原始8,298维向量相同的预测能力，通过药物反应预测。