Arabidopsis enzyme models and one to three rice and maize enzyme models had at least had at least one rhythmic probe. We also have clustered together orthologs and inparalogs of the species we have circadian datasets and only compared those that were common among then. An ortholog catalogue of 18,611 ortholog clusters was created using a combination of InParanoid with MultiParanoid. InParanoid was used to identify orthologs and inparalogs in pairwise proteome comparisons of five different species: 39,021 proteins from sugarcane, 36,338 proteins from sorghum, 106,046 proteins from maize, 51,258 proteins from rice, and 35,386 proteins from Arabidopsis. MultiParanoid was used to merge InParanoid pairwise comparisons into multispecies ortholog clusters. Thus, an ortholog cluster is made of all the orthologs and inparalogs identified among the plant species used. Inparalogs are genes in which duplication happened after the speciation event, in contrast to outparalogs. A caveat of this analysis is that all the paralogs generated from duplication events that occurred after the split from Arabidopsis and the grasses common ancestor were considered inparalogs. However, the incorporation of these paralogs to ortholog clusters LEE011 1211441-98-3 should minimize the problem with inaccurate or incomplete sequences assemblies, a necessity when using data from a species whose genome that has not been completely sequenced like the sugarcane. We have validated the ortholog clusters by comparing the manual annotation of 9 genes associated with sucrose metabolism. Of the 178 annotated genes that were grouped in 28 ortholog clusters, only 2 were false positives and did not correspond to the annotated enzyme model. There were 31 incomplete SAS that were not assigned ortholog clusters but could be annotated manually. Among the other species, there were 17 genes that were false negatives and should be included in an ortholog cluster but were not included in any. The ortholog catalogue allowed us to associate the probes of each array to an ortholog cluster. This way, we can use the probes for ortholog clusters that are common to a pair of arrays, eliminating any selection biases that exist in each array. For example, there were 246 ortholog clusters present 
in both sugarcane dataset and maize dataset. Of these, 114 were considered circadian in sugarcane dataset and 29 were considered circadian in the maize dataset. In total, the sugarcane array had 5,490 ortholog clusters, 2,027 of which were considered circadian, a greater proportion than the other plant datasets. We have tested whether bias could explain the large proportion of rhythmic probes in our dataset. A selection of nine enzyme models associated with sucrose metabolism showed that sugarcane had more enzyme models with at least one rhythmic probe than the other datasets. We also grouped probes into ortholog groups that were common among the compared datasets and sugarcane had more rhythmic ortholog clusters than the compared datasets. Thus, it is possible that the sugarcane circadian clock controls a high proportion of the sugarcane transcriptome. This could be the result of the multiple intra and interspecific hybridizations and intensive selection for desired agronomic traits that increased ploidy and aneuploidy levels in this plant. In particular, the hybridization between S. officinarum and S. spontaneum could result in the enrichment of the circadian controlled genes of each species. It has been observed that patterns of gene expression may drastically change as result of hybridization and increased ploidy in Brassica napus and Arabidopsis thaliana. Ploidy and MK-4827 hybridity effects in the circadian clock have been associated with increase of vigour in Arabidopsis. Comparison among similar datasets from other species showed that there is some overlap between circadian transcripts among different species.