tated transcripts could encode functional proteins. Of the unannotated transcripts, 213 and 436 were differentially expressed in response to salinity stress. These unannotated transcripts encoded proteins inhibitor Volasertib associated with functions such as amino acid metabolism in response to abiotic stress, diterpenoid biosynthesis, and mechanosensitive ion channel function. Mechanosensitive ion channels are gated directly by physi cal stimuli such as osmotic shock and transduce these sti muli into electrical signals. mRNA Seq also captured previously identified genes involved in salinity tolerance, namely those associated with trehalose synthesis, dehydrin, ABA synthesis, sugar transport, glycerol transferase, and transcription factors similar to those of the DREB family.

A substantial number of transcripts were exclusively upregulated only in the root. As only the root was directly exposed to 1 h of salinity stress, it might take time to induce the expression of more genes in the shoot, OsTPP1 might be expressed in the shoot after 10 h of exposure, as has been found in Yukihikari rice. With these genes, Nippon bare may have the potential to be tolerant to salinity stress. Rice cultivars such as Nona Bokra and Pokkali are substantially more salinity tolerant than Nipponbare, suggesting that the genuine salinity stress tolerance gene might be missing in Nipponbare. The 23 Oryza species are geographically, physiologically, and geneti cally diverse, and many of the genes in cultivated rices have been selected by humans under field condi tions, not by environmental stress.

These essentially missing genes could serve as potential genetic resources for the improvement of cultivated crops. Sequence based technology can be used to extract such missing genes by the piling up of short reads on their own gen omes without the need to rely on sequence similarity. Overcoming the technical inaccuracy Microarray technology has been used as a sophisticated platform for the expression profiling of previously anno tated genes. However, as an array based technology, eva luation of signal intensities close to background levels tends to cause artifacts in array analysis because of high levels of background noise and or cross hybridization, moreover, hybridization efficiency might vary with the probes used, suggesting that the calculation of real molar concentrations is inaccurate.

Whereas the Agilent rice 44K Array is designed to quantify 60 mer sequences at the 3 end of transcripts, mRNA Seq quantifies tran script abundance on the basis of Entinostat the number of mapped sequences on the whole gene model. In our study, the two measures of transcript abundance and change ratios were highly correlated, as in a previous report. Moreover, for genes expressed at low or extremely high levels and for genes differentially expressed in arrays, mRNA Seq seemed to be accurate. There fore, mRNA Seq measures the molar concentrations mostly of genes accurately over a broad dynamic range. Biological replication is