DOI: 10.1101/2023.05.22.541678

A variational autoencoder trained with priors from canonical pathways increases the interpretability of transcriptome data

B.Liu B. Rosenhahn T. Illig D. S. DeLuca

摘要

After training five different autoencoder architectures on 22310 transcriptomes, we benchmarked their performance on organ and disease classification tasks on separate selection of 5577 test samples. Every tested architecture succeeded in reducing the transcriptomes to 50 latent dimensions, which captured enough variation for accurate reconstruction. The simple, fully connected autoencoder, performs best across the benchmarks, but lacks the characteristic of having directly interpretable latent dimensions. The beta-weighted, prior-informed VAE implementation is able to solve the benchmarking tasks, and provide semantically accurate latent features equating to biological pathways.

展开全部

图表提取

论文十问

参考文献

被引用

提问与回答

暂无人提供速读十问回答

被引用

笔记

问答

领域