Adversarial generation of gene expression data

Viñas, Ramón; Andrés-Terré, Helena; Lió, Píetro; Bryson, Kevin

doi:10.1101/836254

Cited by 4 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The second method, which we call GAIN-GTEx, is based on Generative Adversarial Imputation Nets (GAIN; Yoon et al, 2018 ). Generative Adversarial Networks have previously been used to synthesize transcriptomics in-silico (Marouf et al, 2020 ; Viñas et al, 2021 ), but to our knowledge their applicability to gene expression imputation is yet to be studied. Similar to generative adversarial networks (GANs; Goodfellow et al, 2014 ), GAIN estimates a generative model via an adversarial process driven by the competition between two players, the generator and the discriminator .…”

Section: Methodsmentioning

confidence: 99%

Deep Learning Enables Fast and Accurate Imputation of Gene Expression

et al. 2021

Self Cite

View full text Add to dashboard Cite

A question of fundamental biological significance is to what extent the expression of a subset of genes can be used to recover the full transcriptome, with important implications for biological discovery and clinical application. To address this challenge, we propose two novel deep learning methods, PMI and GAIN-GTEx, for gene expression imputation. In order to increase the applicability of our approach, we leverage data from GTEx v8, a reference resource that has generated a comprehensive collection of transcriptomes from a diverse set of human tissues. We show that our approaches compare favorably to several standard and state-of-the-art imputation methods in terms of predictive performance and runtime in two case studies and two imputation scenarios. In comparison conducted on the protein-coding genes, PMI attains the highest performance in inductive imputation whereas GAIN-GTEx outperforms the other methods in in-place imputation. Furthermore, our results indicate strong generalization on RNA-Seq data from 3 cancer types across varying levels of missingness. Our work can facilitate a cost-effective integration of large-scale RNA biorepositories into genomic studies of disease, with high applicability across diverse tissue types.

show abstract

Section: Methodsmentioning

confidence: 99%

Deep Learning Enables Fast and Accurate Imputation of Gene Expression

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Specifically, generative adversarial networks (GANs) and Variational Autoencoders (VAE) have been used for generating synthetic WSI and RNA-Seq data. GANs have shown their abilities for modeling cancer characteristics across multiple cancer types, successfully generating synthetic tiles [23, 24] and synthetic gene expression profiles that closely resemble real profiles and capture biological information [25]. VAEs have been successfully applied to gene expression data, showing synthetic generation capabilities in a temporal way and they have been used for data imputation [26, 27].…”

Section: Introductionmentioning

confidence: 99%

RNA-to-image multi-cancer synthesis using cascaded diffusion models

Carrillo-Pérez

Pizurica

Zheng

et al. 2023

Preprint

View full text Add to dashboard Cite

Synthetic data generation offers a solution to the data scarcity problem in biomedicine where data are often expensive or difficult to obtain. By increasing the dataset size, more powerful and generalizable machine learning models can be trained, improving their performance in clinical decision support systems. The generation of synthetic data for cancer diagnosis has been explored in the literature, but typically in the single modality setting (e.g. whole-slide image tiles or RNA-Seq data). Given the success of text-to-image synthesis models for natural images, where one modality is used to generate a related one, we propose RNA-to-image synthesis (RNA-CDM) in a multi-cancer context. First, we trained a variational auto-encoder in order to reduce the dimensions of the patient's gene expression profile, showing that this can accurately differentiate between different cancer types. Then, we trained a cascaded diffusion model to synthesize realistic whole-slide image tiles using the latent representation of the patient's RNA-Seq data. We show that generated tiles preserved the cell-type distribution found in real-world data, with important cell types detectable by a state-of-the-art cell identification model in the synthetic samples. Next, we successfully used this synthetic data to pretrain a multi-cancer classification model, observing an improvement in performance after training from scratch across 5-fold cross validation. Our results demonstrate the potential utility of synthetic data for developing multi-modal machine learning models in data scarce settings, as well as the possibility of imputing missing data modalities by leveraging the information present in available data modalities.

show abstract

“…Similarly, Way et al proposed a VAE trained on pancancer TCGA data, that is able to encode tissue characteristics in the latent space and also leverages biological signals [30]. Recently, Vinas et al presented an adversarial methodology for the generation of synthetic gene expression profiles that closely resemble real profiles and capture biological information [31]. The generation of high-quality WSI tiles has also been researched in recent years given the success of GANs in generating natural images [32,33].…”

Section: Introductionmentioning

confidence: 99%

Synthetic whole-slide image tile generation with gene expression profiles infused deep generative models

Carrillo-Pérez

Pizurica

Ozawa

et al. 2022

Preprint

View full text Add to dashboard Cite

The acquisition of multi-modal biological data for the same sample, such as RNA sequencing and whole slide imaging (WSI), has increased in recent years, enabling studying human biology from multiple angles. However, despite these emerging multi-modal efforts, for the majority of studies only one modality is typically available, mostly due to financial or logistical constraints. Given these difficulties, multi-modal data imputation and multi-modal synthetic data generation are appealing as a solution for the multi-modal data scarcity problem. Currently, most studies focus on generating a single modality (e.g. WSI), without leveraging the information provided by additional data modalities (e.g. gene expression profiles). In this work, we propose an approach to generate WSI tiles by using deep generative models infused with matched gene expression profiles. First, we train a variational autoencoder (VAE) that learns a latent, lower dimensional representation of multi-tissue gene expression profiles. Then, we use this representation to infuse generative adversarial networks (GAN) that generate lung and brain cortex tissue tiles, resulting in a new model that we call RNA-GAN. Tiles generated by RNA-GAN were preferred by expert pathologists in comparison to tiles generated using traditional GANs and in addition, RNA-GAN needs fewer training epochs to generate high-quality tiles. Finally, RNA-GAN was able to generalize to gene expression profiles outside of the training set, showing imputation capabilities. A web-based quiz is available for users to play a game distinguishing real and synthetic tiles: https://rna-gan.stanford.edu/ and the code for RNA-GAN is available here: https://github.com/gevaertlab/RNA-GAN.

show abstract

Adversarial generation of gene expression data

Cited by 4 publications

References 15 publications

Deep Learning Enables Fast and Accurate Imputation of Gene Expression

Deep Learning Enables Fast and Accurate Imputation of Gene Expression

RNA-to-image multi-cancer synthesis using cascaded diffusion models

Synthetic whole-slide image tile generation with gene expression profiles infused deep generative models

Contact Info

Product

Resources

About