PMC:3091640 / 38535-39475 JSONTXT

{"project":"2_test","denotations":[{"id":"20646326-7984417-10480774","span":{"begin":166,"end":168},"obj":"7984417"},{"id":"20646326-17488738-10480775","span":{"begin":326,"end":328},"obj":"17488738"}],"text":"Phylogenetic trees were generated for a group of DNA repair protein homologs. Protein sequences were aligned using the Clustal W multiple sequence alignment program [46]. Only unambiguously aligned positions (excluding poorly conserved and gap regions) were used in phylogenetic analysis, which was performed using the MEGA4 [47]. Neighbor-Joining and Parsimony methods were used for phylogenetic tree searching and inference. The phylogenetic trees were tested by bootstrap analysis with 10000 replications and strict consensus trees were constructed. Similar topologies were found for both algorithms employed, and only Neighbor-Joining being displayed. We used PGDD naming convention for the populus and sorghum dataset in this analysis as the original naming style for predicted models were un-informative. The respective accession numbers used in this analysis and the original naming of predicted genes are given in Additional file 9."}