Venatorbacter cucullus gen. nov sp. nov a novel bacterial predator

A novel Gram-stain negative, aerobic, halotolerant, motile, rod-shaped, predatory bacterium ASxL5T, was isolated from a bovine slurry tank in Nottinghamshire, UK using Campylobacter hyointestinalis as prey. Other Campylobacter species and members of the Enterobacteriaceae were subsequently found to serve as prey. Weak axenic growth on Brain Heart Infusion agar was achieved upon subculture without host cells. The optimal growth conditions were 37 °C, at pH 7. Transmission electron microscopy revealed some highly unusual morphological characteristics related to prey availability. Phylogenetic analyses using 16S rRNA gene sequences showed that the isolate was related to members of the Oceanospirillaceae family but could not be classified clearly as a member of any known genus. Whole genome sequencing of ASxL5T confirmed the relationship to members the Oceanospirillaceae. Database searches revealed that several ASxL5T share 16S rRNA gene sequences with several uncultured bacteria from marine, and terrestrial surface and subsurface water. We propose that strain ASxL5T represents a novel species in a new genus. We propose the name Venatorbacter cucullus gen. nov., sp. nov. with ASxL5T as the type strain.

The genome sequence of ASxL5 T reveals a relationship with marine bacteria. Determination of the 16S rRNA gene sequence (accession number MT636545.1) enabled database searches to establish the www.nature.com/scientificreports/ sequence resembled those in the class Gammaproteobacteria and were most closely aligned with marine bacteria in the family Oceanospirillaceae ( Fig. 2) with members of the genera Thalassolituus and Oceanobacter the nearest relatives. The 16S rRNA gene sequences were notably diverged from predatory bacteria belonging to the family Bdellovibrionaceae (Deltaproteobacteria). The pairwise alignments for B. bacteriovorus HD100 T (type strain, DSM 50701) and B. bacteriovorus DM11A were 48.4% and 47.7% identity, and for B. exovorus JSS 46.7% identity. The ASxL5 T bacterium had 3 copies of the 16S rRNA genes with two being identical to each other and the third differing by 3 bases. Two further predatory bacterial isolates from the same location with similar morphology and phenotypic characteristics (ASx5S and ASx5O; 16S rRNA gene accession numbers MT636546.1 and MT636547.1 respectively) were not identical, but clustered with ASxL5 T and uncultured bacterial database sequences, separate from other genera in the Oceanospirillaceae (Fig. 2). The whole genome sequence of ASxL5 T was determined and deposited in the NCBI database under the accession number CP046056. The genome of ASxL5 T consisted of a single circular chromosome of 2,831,152 bp with a G + C ratio of 56.1%. The genome sequence contained 2653 CDSs (total), of which 2567 were predicted to encode proteins, and of these 1596 could be assigned a putative function (60.2%). The genome contained 67 RNA encoding genes comprising of 9 rRNAs (3 each 5S, 16S and 23S) together with 57 tRNAs. The genomic characteristics of ASxL5 T were compared to the available genomes of the type strains of the closest relatives identified from the 16S rRNA gene sequences ( Table 2). All available Thalassolituus genomes were compared with ASxL5 T using amino acid identity (AAI). The closest available genome sequence (incomplete) determined by AAI was that of Thalassolituus sp. C2-1 (accession NZ_VNIL01000001). This strain was isolated from a deep-sea sediment of the Mariana Trench, but no phenotypic information regarding this strain is available for comparison at present. This organism has a much larger genome at 4.36 Mb compared to 2.82 Mb for ASxL5 T . The average genome size for a member of the order Oceanospirillales is approximately 4.16 Mb (± 1.1; n = 92 complete reference genomes surveyed from https:// www. ncbi. nlm. nih. gov/ assem bly), so the genome of ASxL5 T is quite small compared to other members of the order. A genome-based estimated maximum-likelihood phylogenetic tree (Fig. 3A) was generated with GToTree 1.5.54 utilizing aligned and concatenated amino acid sequences of 172 single-copy genes specific to Gammaproteobacteria [11][12][13][14][15][16][17][18] . This analysis demonstrated a close relationship to Thalassolituus, Bacterioplanes, and Oceanobacter genera. However, these data indicate that ASxL5 T is distinct from its relatives in the Oceanospirillaceae for which genomic sequence data are available. Examination of the component genes present in the ASxL5 T genome using the KEGG database 19 revealed metabolic pathways typical of an aerobic Gammaproteobacterium. ASxL5 T contains a total of 75 genes assigned to bacterial motility proteins, including those involved in chemotaxis, flagella assembly and type IV pilus systems. Within the last category 9 out of 10 genes are responsible for twitching motility in a range of other organisms. The genome of ASxL5 T contained the complete ectoine biosynthesis pathway involved in the protective response to osmotic stress 20 , as might be expected for a halophilic organism. The genome also contains the complete pathways for many cofactors and vitamins including the riboflavin synthesis pathway. Hydrocarbon utilization pathways were incomplete although a gene for alkane 1-monooxygenase (alkB2) was present in ASxL5 T . Homologues of genes identified as largely responsible for hydrocarbon degradation in T. oleiverans MIL-1 T21 such as TOL_2658 (alkB) and TOL_2772 (alcohol dehydrogenase) were notably absent in the genome sequence of ASxL5 T . A comparison of the distribution of genes in COG categories for ASxL5 T with T. oleiverans MIL-1 T is presented in Fig. 3B. Overall, the smaller genome of ASxL5 T contained proportionally less genes from each COG category compared to the larger related genomes. When the number of genes in each functional category are expressed as a percentage of the genome, differences were noted in the percentage of genes in the translation, ribosomal structure, and biogenesis categories together with the energy production and conversion functional categories, which comprise a greater percentage of the ASxL5 T genome than the same groups present in the T. oleiverans MIL-1 T genome. In contrast, T. oleivorans MIL-1 T has a greater percentage of genes within the replication, recombination and repair, and transcription categories compared to the ASxL5 T genome. Interestingly the greatest difference in the contents of each of the functional categories for the two genomes was the number of unknown genes present in ASxL5 T (Fig. 3B). KEGG module enrichment analysis was carried out where each KEGG module represents a collection of manually defined functional units used for annotation and the biological interpretation of genome sequence data. A comparison of the distribution of genes in complete KOG module pathways for ASxL5 T with T. oleiverans MIL-1 T is presented in Fig. 3C. This analysis indicates that while ASxL5 T www.nature.com/scientificreports/ has complete pathways for sulfur and nitrogen metabolism but T. oleiverans MIL-1 T does not. In contrast T. oleiverans MIL-1 T has a complete pathway for cysteine and methionine metabolism that is incomplete in ASxL5 T . Accordingly, ASxL5 T has a signature module (defined as a set of genes that can be used as a phenotypic marker such metabolic capacity or pathogenicity; https:// www. genome. jp/ kegg/ module. html) for sulfate assimilation that is absent in T. oleiverans MIL-1 T . Comparison of the gene content of ASxL5 T with a list of genes suggested to be indicative of a predatory lifestyle 3 was inconclusive. Whilst the waaL gene which encodes a ligase associated with linking O-antigen polysaccharide to the core was present in the ASxL5 T genome (but is common in many Gram-negative bacteria), a tryptophan 2,3-dioxygenase (TDO) gene that may include a 60 amino acid region common in predatory bacteria 22 , was absent. Additional predatory signature genes included those that encode enzymes involved in the mevalonate pathway isoprenoid biosynthesis 3 were also absent from the genome of ASxL5 T . The transcriptional regulator gene gntR was noted to be absent in the predator group examined 3 , but three gntR-like genes could be identified in ASxL5 T .
Phenotypic comparison with related bacteria. The phenotypic characteristics of ASxL5 T are summarised in Table 3 and compared to the phenotypic characteristics reported in the literature for related genera [23][24][25][26][27] . Isolates from T. marinus, T. olevorans, B. sanyensis and Oceanobacter kriegii are motile, halotolerant, oxidase positive rods, but have few other phenotypic characteristics in common with ASxL5 T . The average pH of the oceans is 8.1 (https:// ocean. si. edu/ ocean-life/ inver tebra tes/ ocean-acidi ficat ion# secti on_ 77), a feature reflected in the basic pH survival range of T. marinus, T. olevorans, B. sanyensis and O. kriegii. ASxL5 T is adapted to a greater pH range (4-9) typical of non-marine species. The phenotypic characteristics of Thalassolituus sp. C2-1. are unknown. The growth temperature range of ASxL5 T was generally wider than the marine strains (4-42 °C), although some but not all isolates of T. marinus were thermotolerant. Further phenotypic characterisation was hampered by the inability to grow ASxL5 T in broth medium. Using API 20E tests with material scraped from BA plates, ONPG, arginine dihydrolase, lysine decarboxylase, ornithine decarboxylase, citrate utilisation, urease, tryptophan deaminase, gelatin hydrolase, tests were all negative, whilst indole, acetoin and H 2 S were not produced. Carbohydrates that were not fermented included: glucose, mannose, inositol, sorbitol, rhamnose, sucrose, melibiose, amygdalin and arabinose. The cellular fatty acid profiles of strain ASxL5 T compared to published related reference strains are shown in Table 4. The predominant cellular fatty acids were C16 : 1ω6c and/or C16 : 1ω7c, C16:0, and C18:1 ω9. Hydroxy fatty acids C12:0 3-OH and C10:0 3-OH were also present. The proportion of C16:0 was higher in ASxL5 T than the reported values for related genera. In contrast, there was a reduced proportion of C18 : 1ω7c and/or C18 : 1 ω6c in ASxL5 T compared to that reported for T. marinus IMCC1826 T T. oleivorans MIL-1 T and O. kriegii DSM 6294 T but not detected in B. sanyensis KCTC 32220 T . A comparison to the fatty acid profiles of ASxL5 T with ASxLS revealed minor differences in quantities of individual fatty acids between the two strains, which is consistent with the genomic DNA sequences that they belong to the same species. Poly-3-hydroxybutyrate (PHB) granules were not detected using the Sudan Black test.
ASxL5 T preys on Campylobacter species and other Gram-stain negative hosts. The predatory activity of the ASxL5 T bacterium was investigated to determine the prey range.  Fig. 4A-D whilst the interaction with C. jejuni PT14 and C. hyointestinalis S12 are shown in Fig. 4E-H. The attack mechanism appeared to differ between the prey types tested, with one or more E. coli cells becoming attached to each ASxL5 T cell, positioned laterally along the extended cell before adsorption. In contrast ASxL5 T appeared to attach to campylobacters via a single contact point, often with the predator cell apex, making contact near the Campylobacter cell apex (Fig. 4H).

Discussion
Predatory bacteria have evolved to exploit abundant prey sources. It is becoming apparent that they are widespread in many different environments. The ASxL5 T bacteria was able to be isolated from slurry using phage isolation methods because of the narrow dimensions of members of the population. The genomic relatedness of ASxL5 T to members of the marine bacterial family Oceanospirillaceae was surprising, even though the organism was halotolerant being able to grow on 5% salt containing medium. Water quality analysis of the slurry revealed the sodium chloride level to be less than 0.1%. The slurry is therefore far from a marine environment-geographically and chemically. The presence of three related, but non-identical isolates from the same source, provided evidence that these predators were thriving in this non-marine environment. Moreover, microbiome analysis (data files available from https:// www. ebi. ac. uk/ ena/ brows er/ view/ PRJEB 38990) revealed identical 16S rRNA gene sequences to be in the top 50 most abundant operational taxonomic units (OTUs) in the slurry over several sampling intervals. Several uncultured bacteria were identified in the Genbank database that had similar 16S rRNA gene sequences to the ASxL5 T bacterium. These sequences together with those of ASxL5 T , ASxS5 and ASxO5 appeared to represent a distinct clade separated from Thalassolituus and Oceanobacter (Fig. 2). Three of the uncultured bacteria (GQ921362, GQ921357 and GQ921396) were all isolated from fracture water, from a depth of 1.3 km in a South African gold mine in 2009, while a further two (DQ256320 and DQ337006) were obtained from subsurface water (also in South Africa) in 2005. The most closely related 16S rRNA gene sequence relative to ASxL5 T is a partial 16S rRNA gene sequence that was obtained from enrichment culture of sandy sediment, obtained from a beach in Northern France in 2006 (accession number AM292408 28 ). A further closely related 16S rRNA gene sequence from an uncultured bacterium, HQ183822.1, was obtained from a collection pool leached from a municipal landfill site in China 29 . Clearly the ASxL5 T bacteria is not highly represented in taxonomic databases but it is likely that these sequences from uncultured bacteria represent similar organisms to ASxL5 T , which are distributed worldwide, often in challenging environments. The closest relatives to  [23][24][25][26][27] . Thalassolituus are members of the marine obligate hydrocarbonoclastic bacteria (OHCB) and are prevalent in marine and terrestrial environments often becoming dominant following incidents of hydrocarbon pollution 30,31 . Oceanobacter are not members of the OHCB group but are isolated from marine environments 32 .  www.nature.com/scientificreports/ The phenotypic data indicate that ASxL5 T is a novel species and a member of a previously unrecognised genus within the family Oceanospirillaceae. There are at present no unambiguous criteria for assignment of a newly isolated strain into a new genus. Attempts have been made to identify a universal genus boundary, for example that based on the genomic percentage of conserved proteins (POCP), with a suggested cut off value of 50% identity to reference strains 33 . Others have suggested using AAI values, which have an advantage over POCP in that they can be obtained from incomplete genomes 34 . The authors suggested that a strain is a representative of a different genus if the AAI value is less than 74% when compared with the type strain of the type species. The type genus in the family Oceanospirillaceae, is Oceanospirillum and the type strain is O. linum ATCC 11336 T . The AAI value between ASxL5 T and O. linum ATCC 11336 T is 54.34% and between ASxL5 T and T. oleivorans MIL-1 T (genus type strain) is 67.61% indicating that ASxL5 T represents a novel genus distinct from Thalassolituus. Using the 16S rRNA gene sequences as the taxonomic criteria with a suggested genus delimitation boundary of 94.5% 35 would potentially place ASxL5 T within the genus Thalassolituus exhibiting a 16S rRNA sequence identity of 95.03% with T. oleivorans MIL-1 T and 96.17% to T. marinus IMCC1826 T . However, it would also be placed in the Bacterioplanes genus with a 16S rRNA gene identity with B. sanyensis NV9 of 94.64% illustrating that the use of a single gene such as the16S rRNA gene, can lead to arbitrary taxonomic assignments. Another suggested method uses both ANI and genome alignment fraction (AF) to examine the clustering of data points from all the type and non-type strains of existing genera 36 . The authors suggest the use of a genus demarcation boundary in conjunction with the estimated genus inflection point that is specific to the taxon that is being analyzed. However, without sufficient complete genome sequences from Thalassolituus isolates it is not possible to determine whether ASxL5 T belongs to the Thalassolituus genus by this method. The whole genomic phylogenic tree should be interpreted with discretion due to the limited availability of complete genome sequences to carry out the analysis and secondly methods for whole genome comparison do not account for the substantial difference in the sizes of the genomes being compared. They measure the similarity of core single-copy genes that are conserved between related genera but do not take into account the very large number of genes which are not present in the much smaller genome of ASxL5 T . Clearly ASxL5 T and the group including Thalassolituus, Oceanobacter and Bacterioplanes have a common ancestor, but evolution has taken a different path resulting in genome reduction, possibly as an adaption to a predatory lifestyle. This contrasts with T. oleivorans MIL-1 T that is 28% larger, and which has evolved under different environmental pressures to utilise hydrocarbons 23,30 . An interesting comparison can be made to obligate intracellular parasites and symbionts such as Rickettsia, Chlamydia and Buchnera which have genome sizes of approximately 1 Mb, having undergone significant evolutionary genome degradation as the ability to exploit host-cell metabolites leads to gene loss 37 . An evolutionary change from a marine chemolithotroph to predatory lifestyle could result in a similar genome size reduction. COG and KEGG analysis highlighted global differences in the numbers of genes devoted to specific functions and in the genomic pathways between ASxL5 T and T. oleivorans MIL-1 T and are not due to the extensive acquisition of mobile genetic elements. The difference in the G + C ratios for the whole genomes of ASxL5 T of 56.1% and T. oleivorans MIL-1 T of 46.6% is also indictive of genus separation.
Examination of the coding content of the ASxL5 T genome provided functional insights into the phenotypic characteristics. The presence of genes that encode type IV pili (Tfp) are of particular interest as these facilitate cell movement referred to as social gliding or twitching without flagella over surfaces. Tfp are reported to have other functions including predation, pathogenesis, biofilm formation, natural DNA uptake, auto-aggregation of cells and development 38 . That the ASxL5 T genome contains 18 genes encoding diguanylate cyclase (enzyme that catalyses the conversion of 2 guanosine triphosphate to 2 diphosphate and cyclic di-GMP) and 6 genes encoding the corresponding diguanylate cyclase phosphodiesterase (catalyses the degradation of cyclic di-GMP to guanosine monophosphate) is of interest because cyclic-di-GMP is an important second messenger involved in processes that include biofilm development and detachment, motility, cellular attachment and virulence 39,40 . It should also be noted that in Bdellovibrio bacteriovorus cyclic-di-GMP has been shown to control the switch between free-living and predatory lifestyles 41 .
Most research into predatory bacteria has centred on Bdellovibrio, Bdellovibrio-like organisms and Myxocococcus species. These and other known examples of predatory bacteria form a taxonomically diverse group. Despite this diversity, a group of signature protein families that reflect the phenotypes of 11 known predatory bacteria have been identified 3,22 . However, only the gene encoding O-antigen ligase (waaL) was identified, which is notably prevalent in Gram-negative bacteria. This form of analysis was not helpful for the assignment of ASxL5 T as a predator, likely because of the novel attack strategies it employs. The availability of more diverse predatory bacterial genomes will enable the development of finer resolution analyses that take into account evidence of functional and environmental differences between group members. Examples of predatory bacteria not included in this analysis include Cupriavidus necator 42 and members of the Bradymonabacteria 43 as more predatory taxa are established as researchers investigate different microbial communities.
The most remarkable features of the ASxL5 T bacterium as captured by TEM images, are its unique flexible morphologies that facilitate interactions with prey bacteria. The type of interaction observed is different from other predatory bacteria and has not been identified or reported previously. A proposed predatory life cycle of ASxL5 T is shown in Fig. 5. There are few examples in the literature of similar apical structures to those we report here, but these include those of Terasakiispira papahanaumokuakeensis an Oceanospirillaceae bacterium, that shows occasional apical enlargement 44 , and the Alphaproteobacteria, Terasakiella pusilla previously in the genus Oceanospirillum, that exhibits what are described as "polar membranes" 45 . The presence of coccal forms in older cultures is a frequent observation particularly for bacteria with curved morphology, such as Vibrio, Campylobacter and Helicobacter [46][47][48] , which probably represents a degenerative state. Further work is required to elucidate the precise life cycle of the ASxL5 T bacterium. to determine how it traps and feeds on its prey, and whether its genome encodes bioactive compounds that can be exploited for medicinal or biotechnological purposes. In addition, the description features of the genus, cells are of 1.63 µm in length by 0.37 µm wide when grown on BA or BHI. Colonies on BHI agar are small reaching 2 mm in diameter after 72 h. They are beige, translucent, circular, convex and shiny. Members of the species can use E. coli, Klebsiella spp. Campylobacter spp. and several other Gram-stain negative bacteria as prey.
The type strain ASxL5 T was isolated in Nottinghamshire UK from bovine slurry and is deposited at National Collection of Type Cultures (UK): accession number NCTC 14397 and the Netherlands Culture Collection of Bacteria (NCCB) accession number NCCB 100775. The complete genome sequence of ASxL5 T has been deposited at Genbank under accession CP046056.

Methods
Isolation and phenotypic characterisation of ASx5L T . The ASxL5 T bacterium was isolated from cattle slurry using a technique for phage isolation 9,49 . The slurry was diluted in 1:9 (w/v) in SM buffer (50 mM Tris-HCl [pH 7.5], 0.1 M NaCl, 8 mM MgSO4·7H2O, and 0.01% gelatine; Sigma Aldrich, Gillingham,UK) then incubated at 4 °C for 24 h with slow rotation to elute the predator into the buffer. The suspension was centrifuged at 3000g for 3 min. The supernatant was collected and subjected to a second centrifugation step for 5 min at 13,000g. The supernatant was then passed through a 0.45 µm membrane filter (Minisart; Sartorius, Gottingen, Germany) and a 0.2 µm membrane filter (Minisart) to remove any remaining bacterial cells. ASxL5 T was able to pass through these filters. Soft agar lawns of an isolate of C. hyointestinalis S12, (NCBI accession number CP040464) from the same slurry, were prepared using standard techniques 49 . The filtered slurry was dispensed as 10 µl droplets in triplicate, on each of these host cell plates and allowed to dry. The plates were incubated at 37 °C under microaerobic conditions (5% O 2 , 5% H 2 , 10% CO 2 , and 80% N 2 ) in microaerobic jars for 48 h. Visible plaques obtained were extracted into SM buffer and transferred to fresh lawns of C. hyointestinalis S12 to propagate the lytic organism further. Once it was established that a bacterium was responsible for the lytic plaques rather than a bacteriophage, attempts were made to cultivate the organism independently from the host and characterize it further. Weak growth that improved on subculture was obtained on Brain Heart Infusion Agar (BHI; CM1136, Oxoid, Basingstoke, UK) and Horse Blood Agar No 2 (BA; CM0271 Oxoid) with 5% v/v defibrinated horse blood (TCS Biosciences Lt, Buckingham, UK, added) by aerobic incubation at 37 °C. Antimicrobial sensitivity testing was performed using the disk diffusion methods in accordance with the National Committee for Clinical Standards guidelines 50 . BHI agar incubated aerobically at 37 °C using discs with the following antibiotics (Oxoid): amoxycillin and clavulanic acid 30 µg; cefotaxime 30 µg; streptomycin 10 µg; cip-  Cells were collected in MRD, centrifuged at 13,000g and re-suspended in MRD to an OD 600 of 0.4. An aliquot of 0.5 ml of the suspensions was added to 5 ml aliquots of molten NZCYM top agar (0.6% agar) and poured on to 1.2% NZCYM baseplates. Once set and dried, serial dilutions of ASxL5 T were dispensed as 20 µl droplets in triplicate onto each lawn plate. The incubation temperature and atmosphere were dependent on the test bacteria's requirements.
16S rRNA gene and whole genome sequence determination. DNA was prepared from bacterial isolates using GenElute™ Bacterial Genomic DNA Kit (Sigma Aldridge). PCR amplification of 16S rRNA gene and sequence determination of the product using dye-terminator chemistry (Eurofins Value Read Service, Germany) was carried out using standard methods. The sequences were compared with other 16S rRNA gene sequences using the BLAST-N program to identify and collect closely related species. These were aligned using ClustalW within the MEGA X program 52 . Phylogenetic trees were reconstructed using MEGA X using the Maximum Likelihood method based on the Tamura-Nei model 53 with 1000 bootstrap replications 54 . DNA for whole genome sequencing was extracted using the PureLink™ Genomic DNA Kit (Fisher Scientific, Loughborough, UK). The genome sequence of ASxL5 T was determined using a combination of Illumina MiSeq consisting of 250 bp paired-end reads using libraries prepared from the Nextera tagmentation kit, and long reads of 2 to 20 kb from the PacBio (Pacific Biosciences) platform performed at the Nu-Omics DNA Sequencing Research Facility, Northumbria University. The genome was assembled using CLC Genomics Workbench 12.0.3 (Qiagen, Aarhus, Denmark). ASxL5 T cultures were deposited at National Collection of Type cultures (UK) and the Netherlands Culture Collection of Bacteria (NCCB). Genomes of related organisms used for comparisons were: Thalassolituus oleivorans MIL-1 T (accession HF680312, complete); Bacterioplanes sanyensis KCTC 32220 T (accession BMYY01000001, incomplete); Oceanobacter kriegii DSM 6294 T (accession NZ_AUGV00000000, incomplete); Marinomonas communis DSM 5604 T (accession ASM436330v1, incomplete), Oceanospirullum linum ATCC 11336 T (accession MTSD02000001, incomplete) and Thalassolituus sp. C2-1 (accession NZ_VNIL01000001, incomplete). The alignment fraction (AF) and average nucleic acid identity (ANI) were determined using the JGI Genome Portal 36 hosted at https:// img. jgi. doe. gov// cgi-bin/ mer/ main. cgi? secti on= ANI& page= pairw ise. Amino acid identities (AAI) were determined using the method of Rodriguez-R & Konstantinidis 55 . An estimated maximum-likelihood phylogenetic tree was generated with GToTree 1.5.54 [11][12][13][14][15][16][17][18] . Input genomes representing available reference genomes were selected from reference genera identified as related to ASxL5 T from the 16S rRNA phylogeny. The tree was annotated using the Interactive Tree Of Life online tool, (https:// itol. embl. de/). Functional annotation and analysis of the ASxL5 T genome was carried out using KEGG (Kyoto encyclopedia of genes and genomes) module enrichment assignment using the BlastKOALA KEGG online tool 19 . The distribution of COG categories (clusters of orthologous groups) was determined using the eggNOG-mapper online tool 56 .