The article by Stéphane Audic and Jean-Michel Claverie from the Laboratory of Structural and Genetic Information at the Centre National de la Recherche Scientifique in Marseille, France, focuses on the significance of digital gene expression profiles. They present a systematic study on the influence of random fluctuations and sampling size on the reliability of transcript profiles generated by sequencing thousands of randomly selected cDNA clones. The authors establish a rigorous significance test and demonstrate its use on publicly available transcript profiles. The test links the threshold of selection for putatively regulated genes to the fraction of false-positive clones one is willing to accept. They show that the statistical framework can be used to analyze transcript profiles quantitatively, providing a more precise and extended understanding of the limits within which digital Northern data can be used. The study also compares the test with Fisher’s exact test and discusses the false alarm rate, emphasizing the importance of adjusting the significance threshold based on the number of candidate genes and the acceptable rate of false positives. The results highlight the potential of digital methods for analyzing differential gene expression without the need for repeated or standardized experiments, making them a valuable tool for both basic and pharmaceutical research.The article by Stéphane Audic and Jean-Michel Claverie from the Laboratory of Structural and Genetic Information at the Centre National de la Recherche Scientifique in Marseille, France, focuses on the significance of digital gene expression profiles. They present a systematic study on the influence of random fluctuations and sampling size on the reliability of transcript profiles generated by sequencing thousands of randomly selected cDNA clones. The authors establish a rigorous significance test and demonstrate its use on publicly available transcript profiles. The test links the threshold of selection for putatively regulated genes to the fraction of false-positive clones one is willing to accept. They show that the statistical framework can be used to analyze transcript profiles quantitatively, providing a more precise and extended understanding of the limits within which digital Northern data can be used. The study also compares the test with Fisher’s exact test and discusses the false alarm rate, emphasizing the importance of adjusting the significance threshold based on the number of candidate genes and the acceptable rate of false positives. The results highlight the potential of digital methods for analyzing differential gene expression without the need for repeated or standardized experiments, making them a valuable tool for both basic and pharmaceutical research.