Abiotics stresses such as drought, cold, and salt are very pervasive in modern agriculture, which cause tremendous yield loss each year worldwide. MicroRNAs (miRNAs) play important regulatory roles in abiotic stress responses. To understand the roles of miRNAs in crop responses to abiotic stresses, we used next-generation sequencing techniques to profile the small RNA population in crops with and without stress treatments. We identified many novel as well as stress-regulated miRNAs in: cowpea, rice, maize and sorghum. Specifically, we identified 157 candidate miRNAs in cowpea, 227 in rice, 298 in maize and 407 in sorghum, many of which are novel. We identified 44, 28, 31 and 42 stress-regulated miRNAs in cowpea, rice, maize and sorghum, respectively. Interestingly, majority (81%) of stress-regulated miRNAs in maize were cold stress-regulated, while the majority (79%) in sorghum are drought stress-regulated. We further show that the gain of species-specific miRNA families, rather than family expansion, is responsible for different stress tolerance in maize and sorghum. We provided solid evidence to show that 80 of rice candidate miRNAs originated from MITEs and transposons. Additionally, homology search and structural analysis in collinear regions of maize, sorghum and rice recovered another 26, 10 and 3 miRNAs in maize, sorghum and rice, respectively. We provided clear evidence for orthologous and paralogous miRNA gene expression diversification by northern blot and/or digital expression analysis. Finally, we estimated that the birth rate of miRNAs in cereal genome was 23.55 genes per million years and the death rate was 0.63 and 17.01 genes per million years for orthologous and paralogous miRNA genes, respectively.

The prevalence of large scale genomic studies and technological advances in recent years is promising in the investigation of genome features in more details. The focus of this dissertation is to recruit publicly available genome data and to adopt common and novel computational methods to study the generation and evolution of two classes of small RNAs in cereals. Such a study is important since first small RNAs play a central role in several fundamental processes in cells like genome integrity, gene expression & response to stresses; and second there is a widening gap between the amount of raw and the processed and genomic data which needs to be addressed by computational biology strategies. By the aid of comparative genomics approach the sources of two different types of small RNAs were investigated. Our results provide a strong support that overlapping genes (OGs) could be a source of natural antisense transcripts-small interfering RNAs (nat-siRNA) in cereals while most nat-siRNA generation is not well conserved in them. In addition, our data demonstrate that OGs are common and mostly species specific in maize, rice and Brachypodium and there is no obvious correlation between their number and the total number of genes or genome size; however, genome size and architecture does affect the frequencies and types of overlapping genes. By comparative analysis of the orientation of OGs several birth and death mechanisms were proposed among which translocation and gene creation are the major ones. Moreover, we improved the annotation of microRNA (miRNAs) genes in cereals which produce another type of small RNAs that have crucial regulatory roles in development and stress responses. A novel methodology was developed to use a large number of the recently available RNAseq data to refine gene boundaries of miRNAs and a comparative analysis were performed on them. By defining these upstream regions and using the alignment of the orthologous ones, several miRNA specific regulatory elements were identified which are conserved in cereals and are good candidates for experimental verification. Overall, this dissertation demonstrates that use of publicly available data and computational approaches would increase our understanding of smallRNAs and their evolution. This can provide a foundation for the community to study their expression and function more precisely.