Extracting a biologically relevant latent space from cancer transcriptomes with variational autoencoders

By G. Way, Casey S. Greene. 2017

260 citations recorded by the source corpus.

The Cancer Genome Atlas (TCGA) has profiled over 10,000 tumors across 33 different cancer-types for many genomic features, including gene expression levels. Gene expression measurements capture substantial information about the state of each tumor. Certain classes of deep neural network models are capable of learning a meaningful latent space. Such a latent space could be used to explore and generate hypothetical gene expression profiles under various types of molecular and genetic perturbation. For example, one might wish to use such a model to predict a tumor’s response to specific therapies or to characterize complex gene expression activations existing in differential proportions in different tumors. Variational autoencoders (VAEs) are a deep neural network approach capable of generating meaningful latent spaces for image and text data. In this work, we sought to determine the extent to which a VAE can be trained to model cancer gene expression, and whether or not such a VAE would capture biologically-relevant features. In the following report, we introduce a VAE trained on TCGA pan-cancer RNA-seq data, identify specific patterns in the VAE encoded features, and discuss potential merits of the approach. We name our method “Tybalt” after an instigative, cat-like character who sets a cascading chain of events in motion in Shakespeare’s “Romeo and Juliet”. From a systems biology perspective, Tybalt could one day aid in cancer stratification or predict specific activated expression patterns that would result from genetic changes or treatment effects.

Subfields: Biophysics, Cancer Research, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biophysics

In Cancer Research

In Molecular Biology

Citations in the corpus

Conditional deep surrogate models for stochastic, high-dimensional, and multi-fidelity systems (2019, background): “In a short period of time, this line of work has sparked great interest, and has led to remarkable results in very diverse applications – ranging from the design optimization of light emitting diodes [27], to the design of new molecules [28], to the calibration of cosmological surveys [29], to RNA s”
Deep learning in drug discovery: an integrative review and future challenges (2022, methodology): “As proven by Way and Greene (2018), this is very useful for feature extraction, which condensed a gene expression profile with 5000 dimensions with a maximum of 100 dimensions, some of which revealed to significant characteristics such as the patient’s sexual orientation or melanoma status.”
Applications of machine learning in drug discovery and development (2019, methodology): “Way and Greene 81 introduced a VAE model trained on The Cancer Genome Atlas (TCGA) pan-cancer RNA sequencing data and identified specific patterns in the VAE encoded features.”
Evaluating deep variational autoencoders trained on pan-cancer gene expression (2017, methodology): “Here, we extend a one-hidden layer VAE model, named Tybalt [11].”
Machine learning approaches to drug response prediction: challenges and recent progress (2020, methodology): “Deep learning can certainly help in better understanding cancer biology by predicting binding sites or discovering new biomarkers by analyzing RNA transcripts 47,108,109 .”
Single-cell RNA-seq denoising using a deep count autoencoder (2018, methodology): “Generalizing linear approaches such as principal component analysis, it is commonly used for dimension reduction 23 (see Methods for the detailed description of autoencoders).”
Machine Learning in Drug Discovery: A Review (2021, methodology): “To locate explicit structures in VAE encoded features, then the VAE technique must compete with TCGA (The Cancer Genome Atlas) data in RNA sequencing (Way and Greene 2017).”