Closing Target Trimming: a Perl Package for Discovering Hidden Superfamily Loci in Genomes

Abstract

The contemporary capacity of genome sequence analysis significantly lags behind the rapidly evolving sequencing technologies. Retrieving biological meaningful information from an everincreasing amount of genome data would be significantly beneficial for functional genomic studies. For example, the duplication, organization, evolution, and function of superfamily genes are arguably important in many aspects of life. However, the incompleteness of annotations in many sequenced genomes often results in biased conclusions in comparative genomic studies of superfamilies. Here, we present a Perl software, called Closing Target Trimming, for automatically identifying most, if not all, members of a gene family in any sequenced genomes. Our test data on the F-box gene superfamily showed 78.2 and 79% gene finding accuracies in two well annotated plant genomes, Arabidopsis thaliana and rice, respectively. This annotation performance is clearly higher than the best ab initio methods that are currently available. To further demonstrate the effectiveness of this program, we ran it through 18 plant genomes and five non-plant genomes to compare the expansion of the F-box and the BTB superfamilies. The program discovered that on average 12.7 and 9.3% of the total F-box and BTB members, respectively, are new loci in plant genomes while it only found a small number of new members in vertebrate genomes. Therefore, different evolutionary and regulatory mechanisms of cullin-RING ubiquitin ligases may be present in the plant and the animal kingdoms. Further studies may shed light on new discoveries in the ubiquitin-26S proteasome system-mediated regulatory pathways in eukaryotic organisms. With a detailed compiling instruction and a simple running operation, we expect that this software will assist many biological scientists with little programming experience to smoothly obtain a comprehensive dataset of a gene superfamily in any sequenced eukaryotic genomes.finding_putative_new_loci.pm, Table 1). However, our previous study used BLAT search [29] to locate an annotated gene, which may result in inaccurate genomic coordinates [5].After CTT annotation, we discovered that, on average, 12.7 and 9.3% of the total members in the F-box and BTB families, respectively, are new loci (Tables 3 and 4, S5-S8 Files). Although a slightly higher proportion of F-box genes were discovered than that of BTB genes (p < 0.05, Student's t-test), the percentiles of new members in these two families are significantly correlated (ρ = 0.69, p-value = 0.001, Spearman's correlation test) ( Figure 2B). Therefore, there are various annotation qualities in different sequenced genomes, further highlighting the effectiveness of this package in helping identify most, if not all, superfamily members in a genome for comparative genomic studies. CTT Annotation of the F-box and the BTB Superfamilies in Non-plant GenomesSince the CTT algorithm was first developed to study the F-box gene superfamily in plants [5], we questioned whether this newly designed CTT package was also...

Diversifying Evolution of the Ubiquitin-26S Proteasome System in Brassicaceae and Poaceae

“…For the same reason, RPN11, RPN9B, and regulatory particle triple-A ATPase 1A (RPT1A) were not analyzed. Except for these four members, the remaining 10 RPT members and 14 RPN members from the well characterized Ath RP family were all identified, in addition to the finding of a new RPN1-like protein (AT4G08140; Figure 8A and Figure S3), suggesting the efficiency and accuracy of our highly-automatic CTT annotation program [30]. Within the RP phylogenetic tree, we surprisingly noticed a significant diversity present between the two genera studied.…”

“…Our recent work in developing a highly automatic annotation program, called closing-target-trimming (CTT), allowed us to overcome this annotation limit [30]. Indeed, our previous studies have already applied this gene annotation algorithm to demonstrate that the annotation quality differs significantly among different genome projects, and that CTT can assist in finding most, if not all, members of any gene family in any genome [11,18,30].…”

“…Our recent work in developing a highly automatic annotation program, called closing-target-trimming (CTT), allowed us to overcome this annotation limit [30]. Indeed, our previous studies have already applied this gene annotation algorithm to demonstrate that the annotation quality differs significantly among different genome projects, and that CTT can assist in finding most, if not all, members of any gene family in any genome [11,18,30]. Since each family in the UPS carries a characteristic protein–protein interaction domain feature (Table S1), we used CTT and re-annotated the UPS members, including E1s, E2s, five E3 groups containing seven families, and both the regulatory and core particle families in the 26S proteasome [2,31,32,33] (Figure S1; Section 4).…”

Peer Review #1 of "Adaptive and degenerative evolution of the S-Phase Kinase-Associated Protein 1-Like family in Arabidopsis thaliana (v0.1)"

“…The presence of Skp1 and any additional protein-protein interacting domains in each full-length hit sequence were further confirmed by hmmscan (http://hmmer.org) against the Pfam-A database (Pfam 27, http://pfam.xfam.org). To identify a complete list of Skp1 genes in each species, a previously developed sequence similarity-based annotation algorithm, called Closing Target Trimming (Hua and Early 2018;, was also used to search the genomes for any new Skp1 loci that may not have been annotated.…”

“…halleri, respectively (Supplemental File S1). After a subsequent sequence-similarity based Closing Target Trimming search (Hua and Early 2018;, no additional hits were identified in each genome. The relatively long length of the Skp1 domain, the low number or absence of introns in the Skp1 loci, and the small size of the Skp1 family, may facilitate the annotation of Skp1 genes in genomes.…”