Table 5 reports the Unigene clusters candidate to encode miRNA coding genes on the basis of check this the precursor sequence secondary structure and of the presence of the miRNA. It cannot be excluded that the clusters unable to fold with a miRNA like structure are false negatives for several reasons, such as truncated precursor sequences in EST database. Putative microRNA sequences have also been BLASTed against previously known precursors available from mirBASE, the analysis found similarities with 6 different miRNA families. The secondary struc tures of the putative microRNA precursors are reported in the additional file 4. Linking together sequences con taining miRNA precursors from Dryanova et al. and from the present work, information on several micro RNA putative secondary structures, belonging to 10 miRNA families are now available.
The mature miR NAs predicted from these data are 18 to 24 nt long, with a higher frequency for 20 and 21 nt. Genetic variation at miRNA target sites A single nucleotide change in the sequence of a target site can affect miRNA regulation, as a consequence naturally occurring SNPs in target sites are candidates for relevant functional variations. Nair et al. established a perfect association between a SNP at the miR172 tar geting site and cleistogamy in barley. Overall few papers have been published to date describing variations among plant genotypes at miRNAs and their target sites, while plenty of information is available for humans. Genome wide studies in humans have shown that the levels of polymorphism at miRNA and miRNA target sites are lower than at coding or neutral regions, however beneficial miRNA target site polymorphisms also exist.
In this study, publicly available SNP data have been analyzed in context with miRNAs and their target sites. EST derived SNPs can provide a rich source of biologi cally useful genetic variation due to the redundancy of gene sequence, the diversity of genotypes present in the databases and the fact that each putative polymorphism is associated with an expressed gene. Variations both in functional regions of putative miRNAs and at miRNA target sites have been found. Previous works in human have highlighted a relatively low level of variation in functional microRNA regions and an appreciable level of variation at target sites. Hv.
5064, the candidate for miR1137 coding sequence, has been tested for modifications of pre miRNA struc ture due to a base substitution in position 13. To evaluate the possible impact of this SNP on pre miRNA secondary structure, Gibbs free energy and MFEI from each version of pre miRNA were calculated using mfold program. GSK-3 Data in figure 3 show the structural variation obtained when moving from C variant to G variant with a higher MFEI for the second one and thus a greater stability of the molecule. Difference in G moving from C to G and vice versa were calculated according to Ehrenreich and Purugganan.