Annotation of the sequenced genomes was performed using DOGMA [55] (link), coupled with manual selections for start and stop codons and for intron/exon boundaries. We calculated the average cp genome size of subfamilies in Poaceae on the basis of the species listed in Table 3 . We estimated the monocot mean cp genome size based on Acorus calamus (AJ879453) [56] (link), Dioscorea elephantipes (EF380353) [57] (link), Lemna minor (DQ400350) [58] (link), Oncidium Gower Ramsey (GQ324949) [59] (link), Phalaenopsis aphrodite (AY916449) [60] (link), Phoenix dactylifera (GU811709), and Typha latifolia (GU195652) [61] (link). The rpoC2 sequences from the grass family were aligned to the rpoC2 sequences of tobacco by MEGA 4.0 to determine the insertion size in the gene. We downloaded B. oldhamii and D. latiflorus cp genomes sequences from GenBank, and multiple alignments of eight bamboos cp genomes were made using MAFFT version 5 [62] (link). Full alignments with annotations were visualized using the VISTA viewer. The genetic divergence represented by p-distance was calculated by MEGA 4.0 with species of Arundinarieae as one group and those of Bambuseae as another.
We determined the three types of repeats, dispersed, tandem and palindromic, by first applying the program REPuter and then manually filtering the redundant output of REPuter. Gap size between palindromic repeats was restricted to a maximal length of 3 kb. Overlapping repeats were merged into one repeat motif whenever possible. A given region in the genome was designated as only one repeat type, and tandem repeat was prior to dispersed repeat if one repeat motif could be identified as both tandem and dispersed repeats. For coding, each repeat present in a given genome was ‘1’ and those absent were labeled as ‘0’. We performed MP analyses of this matrix using PAUP*4.0b10 [63] to implement exhaustive tree searches. Non-parametric bootstrap analysis was conducted under 1,000 replicates with TBR branch swapping.
We determined the three types of repeats, dispersed, tandem and palindromic, by first applying the program REPuter and then manually filtering the redundant output of REPuter. Gap size between palindromic repeats was restricted to a maximal length of 3 kb. Overlapping repeats were merged into one repeat motif whenever possible. A given region in the genome was designated as only one repeat type, and tandem repeat was prior to dispersed repeat if one repeat motif could be identified as both tandem and dispersed repeats. For coding, each repeat present in a given genome was ‘1’ and those absent were labeled as ‘0’. We performed MP analyses of this matrix using PAUP*4.0b10 [63] to implement exhaustive tree searches. Non-parametric bootstrap analysis was conducted under 1,000 replicates with TBR branch swapping.
Free full text: Click here
