Transforming L1000 profiles to RNA-seq-like profiles with deep learning

Abstract

The L1000 technology, a cost-effective high-throughput transcriptomics technology, has been applied to profile a collection of human cell lines for their gene expression response to > 30,000 chemical and genetic perturbations. In total, there are currently over 3 million available L1000 profiles. Such a dataset is invaluable for the discovery of drug and target candidates and for inferring mechanisms of action for small molecules. The L1000 assay only measures the mRNA expression of 978 landmark genes while 11,350 additional genes are computationally reliably inferred. The lack of full genome coverage limits knowledge discovery for half of the human protein coding genes, and the potential for integration with other transcriptomics profiling data. Here we present a Deep Learning two-step model that transforms L1000 profiles to RNA-seq-like profiles. The input to the model are the measured 978 landmark genes while the output is a vector of 23,614 RNA-seq-like gene expression profiles. The model first transforms the landmark genes into RNA-seq-like 978 gene profiles using a modified CycleGAN model applied to unpaired data. The transformed 978 RNA-seq-like landmark genes are then extrapolated into the full genome space with a fully connected neural network model. The two-step model achieves 0.914 Pearson’s correlation coefficients and 1.167 root mean square errors when tested on a published paired L1000/RNA-seq dataset produced by the LINCS and GTEx programs. The processed RNA-seq-like profiles are made available for download, signature search, and gene centric reverse search with unique case studies.

In Silico Generation of Gene Expression profiles using Diffusion Models

“…Subsequently, Chen et al . (2016) and Jeon et al . (2022) proposed neural learning approaches using the dimensionality reduction permitted by the L1000 genes to achieve gene expression inference via multilayer perceptrons (MLPs) architectures.…”

“…Building upon the methodology used to deploy and assess data augmentation with VAEs and GANs (Lacan et al, 2023), the proposed contribution is a gene expression generation pipeline leveraging the power of diffusion models. The critical issues related to the high dimensionality of transcriptomics data are handled by considering the so-called 1,000 landmark genes (Subramanian et al, 2017): the data generation is conducted in this reduced 1,000-dimension space, and the generated samples are mapped to the remaining genes space using a trained linear or non-linear approach (Chen et al, 2016;Jeon et al, 2022). To the best of our knowledge, this is the first success in applying the famed diffusion models on bulk RNA data.…”

Text-mining-based feature selection for anticancer drug response prediction

“…The top 500 genes show the highest correlation with drug response were selected as features for training the machine learning model. Landmark-1000 (L1000) gene set is known to be reproducible and capable of inferring expression levels of the majority of other genes ( Subramanian et al 2017 , Jeon et al 2022 ). This gene set is frequently used for characterizing biological samples ( Malta et al 2018 , Wan et al 2020 ) and machine learning-based drug response prediction ( Gardiner et al 2020 , Lu et al 2021 , Uner et al 2023 ).…”

Proteomic Analysis of Huntington’s Disease Medium Spiny Neurons Identifies Alterations in Lipid Droplets

“…Enrichment analysis for GO biological processes with differentially expressed proteins (FDR < 0.05, logFC > 0.58) was done utilizing the R package clusterProfiler. Drug prediction was done utilizing the LINCS L1000 characteristic direction signatures search engine ( https://maayanlab.cloud/L1000CDS2/#/index ) with upregulated and downregulated proteins as input ( 52 ).…”