RNA-seq profiles of various developmental stages and systematic nuclear hormone receptor (NHR) gene knockdowns in C. elegans

The transcriptome is one of the most informative molecular phenotypes and is most frequently measured by RNA-sequencing (RNA-seq). Recently, Perturb-seq has been developed to measure the transcriptome in single cells upon perturbation of different genes by CRISPR-Cas9. While powerful, Perturb-seq has limited resolution and cannot easily be applied to intact animals. Here, we present 'Worm Perturb-Seq (WPS)', which combines whole-animal gene knockdown by RNAi in Caenorhabditis elegans with a large-scale, high-throughput bulk RNA-seq approach that produces high-resolution transcriptome profiles. WPS includes an analytical framework that identifies differentially expressed genes (DEGs) by using gene-specific models with empirical null distributions, rather than using control RNAi alone. Specifically, our approach identifies and corrects inflated P values obtained with standard methods, thus improving statistical rigor. We applied WPS to collect a detailed time course dataset for the animal at various developmental stages, from L2 larve to gravid adult stages. We also applied WPS to 103 C. elegans nuclear hormone receptors (NHRs) and revealed diverse functional modules featuring pairs of NHRs regulating shared target genes. WPS enables large-scale and high-quality transcriptomic analysis of genetic perturbations in an animal and can be implemented not only for RNAi but also for other perturbing conditions such as different bacterial diets, C. elegans strains, etc. WPS may also provide a platform for similar studies in other model organisms and human tissue culture cells. Overall design: In this study, we applied Worm Perturb-seq (WPS), a massively parallel RNAi and RNA-seq technology, to perturb 103 nuclear hormone receptor (NHR) genes in nematode C. elegans, followed by a whole-animal, bulk RNA-seq readout. Samples were multiplexed to form pooled sequencing libraries with around 45 samples per library. The dataset includes eight libraries corresponding to 353 samples, covering the perturbations of 103 genes in biological triplicates. This record also contains the WPS library for time-course samples of C. elegans development that were used in the study. The developmental stage dataset includes 17 time points for animals at L2, L3, L4, Young Adult and Gravid Adult stages, ranging from 17 to 69 hours post L1 seeding, in biological triplicate. All RNA-seq libraries were constructed following WPS method, an adapted CEL-seq2 protocol, giving a multiplexed bulk RNA-seq library for 3'-end sequencing of C. elegans genes. The read 1 is used to encode barcode information and read 2 maps to the 3' end region of C. elegans transcripts. Barcode information can be found from the Supplementary Table "barcode_and_metadata_full_dataset.csv" ("barcode_and_metadata_dev_stages.csv" for developmental stages). These libraries can be directly processed by the WPS data processing pipeline (https://github.com/XuhangLi/WPS) providing the barcode information attached. A custom processing of the data is feasible following the typical demultiplexing methods for single-cell RNA-seq data that are able to process CEL-seq2 library. Please refer to our manuscript for decription of the design of read 1 and the pooling of samples.