Simulated ONT direct RNA and 1D cDNA sequencing data of varying sequencing depths (0.5 million, 1 million, 3 million, and 5 million simulated reads) used for benchmark evaluations of transcript discovery and quantification in our paper "ESPRESSO: Robust discovery and quantification of transcript isoforms from error-prone long-read RNA-seq data". All details can be found in the Materials and Methods section of the paper. HEK293T_DirectRNA.transcriptome_quantification.tsv and HEK293T_DirectRNA.transcriptome_quantification.tsv are tab-separated files containing estimated raw read counts and normalized abundance values (in TPM) of transcripts annotated in GENCODE v34lift37. Transcript quantification was done using NanoSim (version 3.1.0). HEK293T_DirectRNA.NanoSim_500k.fastq.gz, HEK293T_DirectRNA.NanoSim_1M.fastq.gz, HEK293T_DirectRNA.NanoSim_3M.fastq.gz, and HEK293T_DirectRNA.NanoSim_5M.fastq.gz are gzip compressed FASTQ files containing 0.5 million, 1 million, 3 million, and 5 million simulated ONT direct RNA sequencing reads respectively. HEK293T_1DcDNA.NanoSim_500k.fastq.gz, HEK293T_1DcDNA.NanoSim_1M.fastq.gz, HEK293T_1DcDNA.NanoSim_3M.fastq.gz, and HEK293T_1DcDNA.NanoSim_5M.fastq.gz are gzip compressed FASTQ files containing 0.5 million, 1 million, 3 million, and 5 million simulated ONT 1D cDNA sequencing reads respectively.