The paper introduces Cuffdiff 2, an algorithm for differential analysis of gene and transcript expression using high-throughput RNA sequencing (RNA-seq). Cuffdiff 2 addresses the challenges of estimating expression at transcript-level resolution and controlling for variability across replicate libraries. The algorithm robustly identifies differentially expressed transcripts and genes, revealing differential splicing and promoter-preference changes. The authors demonstrate the accuracy of Cuffdiff 2 through a study of lung fibroblasts in response to the loss of the developmental transcription factor *HOXA1*. They show that *HOXA1* is essential for lung fibroblast and HeLa cell cycle progression, and that its loss results in significant changes in thousands of individual transcripts and isoform switching events in key regulators of the cell cycle. Cuffdiff 2 performs robust differential analysis in RNA-seq experiments at transcript resolution, providing a layer of regulation not easily observable with other high-throughput technologies. The algorithm is accurate over a wide range of RNA-seq designs, including those performed on benchtop sequencers like the Illumina MiSeq. The authors also validate Cuffdiff 2's effectiveness in analyzing published RNA-seq data with matched quantitative PCR measurements and demonstrate its performance in simulated data scenarios. Finally, they show that Cuffdiff 2 can recover transcriptome dynamics from the HOXA1 knockdown experiment with less data than the HiSeq 2000, highlighting its potential for broader application in differential analysis.The paper introduces Cuffdiff 2, an algorithm for differential analysis of gene and transcript expression using high-throughput RNA sequencing (RNA-seq). Cuffdiff 2 addresses the challenges of estimating expression at transcript-level resolution and controlling for variability across replicate libraries. The algorithm robustly identifies differentially expressed transcripts and genes, revealing differential splicing and promoter-preference changes. The authors demonstrate the accuracy of Cuffdiff 2 through a study of lung fibroblasts in response to the loss of the developmental transcription factor *HOXA1*. They show that *HOXA1* is essential for lung fibroblast and HeLa cell cycle progression, and that its loss results in significant changes in thousands of individual transcripts and isoform switching events in key regulators of the cell cycle. Cuffdiff 2 performs robust differential analysis in RNA-seq experiments at transcript resolution, providing a layer of regulation not easily observable with other high-throughput technologies. The algorithm is accurate over a wide range of RNA-seq designs, including those performed on benchtop sequencers like the Illumina MiSeq. The authors also validate Cuffdiff 2's effectiveness in analyzing published RNA-seq data with matched quantitative PCR measurements and demonstrate its performance in simulated data scenarios. Finally, they show that Cuffdiff 2 can recover transcriptome dynamics from the HOXA1 knockdown experiment with less data than the HiSeq 2000, highlighting its potential for broader application in differential analysis.