Genome sequences were downloaded from the NCBI for 342 named Pseudomonas species and 46 sequences from other Pseudomonadaceae genera available as of December 16, 2022, in the database. Each species is represented in the tree by a single genomic sequence, which is generally of the type strain, when available. Based on these genome sequences, a rooted phylogenetic tree was constructed based on concatenated sequences of 118 conserved proteins that are a part of the phyloeco set for the class Gammaproteobacteria (Wang and Wu, 2013) (listed in Supplementary Table S1). Genome sequences for Moraxella bovoculi and M. bovis were included in this dataset for rooting purposes. Another comprehensive phylogenetic tree was constructed based on the core proteins from the genomes of Pseudomonadaceae species. This latter tree was based on genome sequences for 174 species, which included most of the species from the other main clades of Pseudomonas species, but only 41 divergent species from the Fluorescens superclade (lineage). Trees were constructed using an internally developed pipeline described in earlier work (Adeolu et al., 2016; Gupta et al., 2020; Rudra and Gupta, 2021; Saini and Gupta, 2021). Briefly, the CD-HIT program (Li and Godzik, 2006; Fu et al., 2012) was used to identify protein families (or homologs of different proteins) where the proteins were present in at least 80% of the genomes in the dataset and they shared at least 50% of sequence length and identity. The Clustal Omega program (Sievers et al., 2011) was then used to generate multiple sequence alignments (MSA) of the proteins. These MSAs were converted into profile Hidden Markov Models (HMMs) using HMMer 3-1b2 (Eddy, 2011), which were then used to search for other members of the protein families in the input genomes. These analyses identified 1,503 protein families meeting the stated criteria (also listed in Supplementary Table S1). The sequence alignments of these proteins were trimmed using TrimAl program (Capella-Gutiérrez et al., 2009) to remove poorly aligned sections prior to their concatenation. The concatenated sequence alignment for the phyloeco set of proteins for Gammaproteobacteria was created similarly using the published profile HMMs for these proteins (Wang and Wu, 2013). The concatenated sequence alignments used for the construction of phyloeco and the core genome trees consisted of 42,362 and 494,143 amino acid (aa) positions, respectively. Using these alignments, maximum likelihood (ML) trees were initially constructed using FastTree 2 (Price et al., 2010) with the Whelan and Goldman (2001) model of protein sequence evolution. The resulting trees were optimized with RAxML 8 (Stamatakis, 2014) and to obtain the Shimodaira-Hasegawa (SH) statistical support values, which are similar to the bootstrap scores, for different nodes. The trees were labeled and formatted using MEGA X (Kumar et al., 2018). The percentage of conserved proteins (POCP) and average amino acid identity (AAI) for different pairs of genomes were calculated as described by Thompson et al. (2013) and Qin et al. (2014).

Identification of CSIs was carried out by similar procedures as described in earlier work (Gupta, 2014, 2016; Gupta et al., 2020). Briefly, local BLASTp searches were carried out on protein sequences from the genomes of several Pseudomonas species representing different clades of interest and other outgroup species. Based on these BLAST searches, sequences of high scoring homologs (E value <1e-20) of different proteins were retrieved for several species (generally between 4 to 12) from the group of interest, and 10–15 species from other Pseudomonas clades or other Pseudomonadaceae genera. Multiple sequence alignments for the proteins were created using Clustal X 2.1 program (Jeanmougin et al., 1998). Alignments were visually examined for insertions or deletions of fixed length that were present in conserved regions (i.e., flanked on both sides by minimally 5–6 conserved aa residues in the neighboring 40–50 aa), and which were only found in the Pseudomonas species from the clade of interest. The indels which were not present in conserved regions were not further considered. The query sequences consisting of the conserved indels and their flanking 30–40 aa on each side were subjected to a second BLASTp search against the NCBI nr database and the top 250–500 hits were evaluated to determine the group specificities of the CSIs. Based on these results, indels which were specific for different clades of Pseudomonas were formatted using the SIG_CREATE and SIG_STYLE programs (Gupta, 2014, 2016). Due to space constraints, sequence information is shown for only a limited number of species in the main figures. However, unless otherwise indicated the CSIs reported here are specifically found in different named Pseudomonas species from the indicated groups. More detailed information for different CSIs is provided in the Supplemental Data files.

To understand the interspecies relationships among different Pseudomonadaceae species whose genomes were available in the NCBI as of December 16, 2022, two genome-scale phylogenetic trees were constructed. The first of these trees shown in Figure 1 (Supplementary Figure S1), which will be referred to as the phyloeco tree, is based on concatenated sequences for 118 conserved proteins, which comprise the phyloeco set for the class Gammaproteobacteria (Wang and Wu, 2013). Another comprehensive tree constructed is a core genome (protein) tree based on 1,503 proteins which are shared by at least 80% of the input Pseudomonadaceae species. This latter tree included only representative species (41) from the Fluorescens superclade (lineage), which is not the focus of this study. In both constructed trees, most observed nodes are supported with 100% SH values (like bootstrap scores) indicating that the observed evolutionary relationships are reliable.

The overall branching and grouping of Pseudomonadaceae species in different clusters in both the phyloeco (Figure 1) and the core protein tree (Supplementary Figure S2) is nearly identical, and it is similar to that observed in our earlier work (Rudra and Gupta, 2021), and other phylogenetic studies (Gomila et al., 2015; Hesse et al., 2018; Peix et al., 2018; Lalucat et al., 2020; Girard et al., 2021; Lalucat et al., 2022; Passarelli-Araujo et al., 2022). In both these trees, Pseudomonas species formed several distinct clades/groups, and species from the genera Azomonas and Azotobacter consistently branched between them (Hesse et al., 2018; Rudra and Gupta, 2021; Lalucat et al., 2022; Passarelli-Araujo et al., 2022). Additionally, species from the two recently proposed genera Stutzerimonas and Chryseomonas also branched within other Pseudomonas species, thus further contributing to the polyphyly of this genus. We have labeled different Pseudomonas species clades in Figure 1 and Supplementary Figure S2 by their commonly used clade/group names (Hesse et al., 2018; Girard et al., 2021; Lalucat et al., 2022). One distinct clade observed in all constructed trees is the “Aeruginosa clade,” which contains the type species P. aeruginosa and 13 other Pseudomonas species. As this clade contains the type species of the genus Pseudomonas, we have labeled it as the “Genus Pseudomonas sensu stricto.” Other species’ clades observed and labeled in Figure 1 (Supplementary Figure S2) include: the Alcaligenes, Anguilliseptica, Azomonas, Azotobacter, Flexibilis, Fluvialis, Linyingensis, Oleovorans, Oryzihabitans, Resinovorans, Straminea, Stutzeri (Stutzerimonas), Thermotolerans, and Fluorescens superclade (lineage).The Genome Taxonomy Database (GTDB),² based on phylogenetic analysis of 120 ubiquitously conserved proteins, now provides an important resource for taxonomic inferences (Parks et al., 2018). The GTDB refers to the “Aeruginosa clade” as the genus Pseudomonas whereas most of the other observed species clades are referred to as distinct genera denoted by designations such as g_Pseudomonas_B, g_Pseudomonas_K, etc., which are also indicated in the tree in Figure 1.

Of these observed clades, the Fluorescens superclade (lineage) is the largest harboring 245 Pseudomonas species. It is separated from all other Pseudomonas species by a long branch in both constructed trees (Figure 1; Supplementary Figure S2). Due to the large number of species present in this clade, it is shown in a compressed form in Figure 1. However, detailed information for species comprising this clade is provided in Supplementary Figure S1. The Fluorescens superclade (lineage) is made up of multiple distinct clades and subclades (see Supplementary Figure S1) (Hesse et al., 2018; Peix et al., 2018; Lalucat et al., 2020; Rudra and Gupta, 2021; Lalucat et al., 2022). However, all species grouping within the Fluorescens superclade (lineage) are part of the GTDB taxon “g_Pseudomonas_E.” Although the Pseudomonas_E cluster in GTDB also encompasses the Alcaligenes, Anguilliseptica, Oleovorans and Thermotolerans clades, these clades in our phylogenomic trees (Figure 1; Supplementary Figure S1), and in several other published studies (Hesse et al., 2018; Girard et al., 2021; Lalucat et al., 2022; Passarelli-Araujo et al., 2022), branch separately from the Fluorescens superclade. This discrepancy in the branching positions of the Alcaligenes, Anguilliseptica, Oleovorans and Thermotolerans clades between the GTDB taxonomy and other phylogenomic trees, was also noted by Lalucat et al. (2022). However, in the present work, we will not be examining the evolutionary relationships of different species within the Fluorescens superclade. Besides the “Aeruginosa clade” and the Fluorescens superclade (lineage), the other clades marked in Figure 1 (Supplementary Figure S2) contain between 2–18 species. Except for the Anguilliseptica clade, which shows poor resolution and weak statistical support, all other clades in our phylogenetic trees are statistically strongly supported. Besides these species’ clades, a limited number of Pseudomonas species (viz. P. indica, P. kuykendallii, P. mangiferae, P. mangrovi, P. matsuisoli and P. pohangensis) are not part of any of the observed clades.

The analyzed genome sequences were also used for determination of percentage of conserved proteins (POCP) and average amino acid identity (AAI) between different pairs of genomes. The results of pairwise AAI and POCP values, for different Pseudomonadaceae genomes are presented in Supplementary Tables S2 and S3, respectively. Genome pairs exhibiting higher AAI or POCP values are shown by a darker shade of green/red, and different clades observed in our phylogenetic trees (Figure 1; Supplementary Figure S2) are outlined. In Table 1, we present a summary of the ranges of the AAI and POCP values for different Pseudomonas species clades for the ingroup and outgroup species. Based on the results in Table 1, the AAI and POCP values for species within different clades are higher (AAI values range: 0.70–1.00; POCP values range: 0.66–1.00) in comparison to these values for species from the other clades (AAI values range: 0.67–0.81; POCP values range: 0.42–0.77), which is an expected result. However, based on the AAI and POCP values (Table 1), only species from the Alcaligenes, Azotobacter, Flexibilis, Fluvialis, Lingyingensis, Oleovorans and Thermotolerans clades show no overlap with species from the other clades. In contrast, these values for several other clades (viz. “Aeruginosa.” Anguiliiseptica, Azomonas, Oryzihabitans, Resinovorans, Straminea, Stutzeri) either show significant overlap or are very close to those from the outgroup species. Thus, based on these genome similarity indices, species from different observed Pseudomonadaceae clades cannot be reliably demarcated. In Table 1, the highest overlap in the AAI and POCP values between the ingroup versus outgroup species is observed for the species from Anguilliseptica clade, which also shows poor resolution and weak statistical support in the phylogenetic trees.

Although Pseudomonadaceae species form similar clades in different genome scale trees (Hesse et al., 2018; Parks et al., 2018; Girard et al., 2021; Lalucat et al., 2022; Figure 1; Supplementary Figure S2), branching of species in phylogenetic trees is influenced by large numbers of variables (Gupta, 1998; Baldauf, 2003; Felsenstein, 2004). Moreover, in phylogenetic trees for Pseudomonas, species from several clades are separated from each other by short branches (Figure 1; Supplementary Figure S2), which makes it difficult to reliably determine their boundaries. The POCP and AAI values for several clades also overlap or are very close to the other species (Table 1), thus they do not permit reliable determination of the boundaries of these clades. Hence, it was important to discover other reliable means for the demarcation of these clades. Molecular synapomorphies consisting of CSIs in genes/proteins sequences, which are uniquely shared characteristics of species from different clades, provide important means for the demarcation of taxa of different ranks in molecular terms (Gupta, 2014; Adeolu et al., 2016; Gupta et al., 2020; Patel and Gupta, 2020; Rudra and Gupta, 2021). Hence, detailed studies were conducted on protein sequences from Pseudomonadaceae species to identify CSIs which are specific for different observed clades. These analyses have identified 98 novel CSIs which are specific for different Pseudomonadaceae clades, providing independent evidence for the genetic distinctness of these clades and affording reliable means for their demarcation. Brief descriptions of the characteristics of these CSIs are given below.

The “Aeruginosa clade” representing the genus Pseudomonas sensu stricto, encompasses 14 named species (viz., P. aeruginosa, P. paraeruginosa, P. citronellolis, P. delhiensis, P. humi, P. jinjuensis, P. knackmussii, P. multiresinivorans, P. nicosulfuronedens, P. nitritireducens, P. nitroreducens, P. panipatensis, “P. pseudonitroreducens” and P. schmalbachii) (Figure 1). Our analyses have identified six CSIs in proteins involved in different functions (Table 2), which are commonly and, in most cases, uniquely shared by different species from the “Aeruginosa clade.” Sequence information for one of these is presented in Figure 2. In the example shown, a two aa insertion (highlighted) in a conserved region of the HugZ family protein is commonly shared by all 14 species from the “Aeruginosa clade” but absent in all other Pseudomonadaceae species. Sequence information is shown in Figure 2 for only a limited number of species. However, more detailed information for this CSI is presented in Supplementary Figure S3. Like the CSI shown in Figure 2, we have identified five additional CSIs in other proteins which, except for an isolated occurrence, are uniquely shared by different species from the “Aeruginosa clade.” Sequence information for these CSIs is provided in Supplementary Figures S4–S8 and some of their characteristics are summarized in Table 2. Due to their unique shared presence in species from the “Aeruginosa clade,” genetic changes responsible for these CSIs likely occurred in a common ancestor of this clade and subsequently inherited by all members. Due to their specificities for the species from the “Aeruginosa clade,” these molecular synapomorphies provide robust means for the demarcation of this clade in molecular terms.

P. alcaligenes was indicated to branch separately from other clades in earlier studies (Hesse et al., 2018; Girard et al., 2021; Lalucat et al., 2022). In our phylogenetic trees (Figure 1; Supplementary Figure S2), three recently identified species (viz., P. campi, P. guryensis, P. ullengensis) also reliably grouped with P. alcaligenes. Our analysis has identified six novel CSIs, which in most cases are exclusively shared by all four species from the Alcaligenes clade. Sequence information for one of these CSIs is presented in Figure 3A, where a two aa insertion in the protein ferric iron uptake transcriptional regulator is exclusively present in all four species from the Alcaligenes clade. Five additional CSIs in other proteins are also generally specific for the species from this clade. Detailed sequence information for these six CSIs is provided in Supplementary Figures S9–S14, and some of their characteristics are listed in Table 2. The identified CSIs provide reliable means for the demarcation of species from the Alcaligenes clade in molecular terms and we are proposing their transfer into Aquipseudomonas gen. nov.

Oleovorans clade is a strongly supported clade consisting of 15 Pseudomonas species (viz., P. alcaliphila, P. chengduensis, P. composti, P. guguanensis, P. hydrolytica, “P. indoloxydans,” P. khazarica, P. mendocina, P. oleovorans, P. pseudoalcaligenes, “P. sediminis,” “P. sihuiensis,” P. toyotomiensis, “P. wenzhouensis,” P. yangonensis), which reliably group together in the constructed phylogenetic trees (Figure 1; Supplementary Figure S2). The genetic distinctness of this clade is also independently supported by five novel identified CSIs which, excepting an isolated occurrence, are uniquely shared by all species from this clade. Sequence information for one of these CSIs is provided in Figure 3B, where a one aa deletion (highlighted), within a conserved region of the protein cysteine synthase A, is exclusively shared by all species from the Oleovorans clade. More detailed sequence information for this CSI and four additional CSIs specific for the Oleovorans clade is provided in Supplementary Figures S15–S19 and some of their characteristics are listed in Table 2. Based on the strong evidence presented here demonstrating the distinctness of species from the Oleovorans clade, we are proposing the transfer of these species into Ectopseudomonas gen. nov.

In addition to the species with validly published names, Oleovorans clade also encompasses four species [viz., “P. indoloxydans” (Manickam et al., 2008), “P. sediminis” (Behera et al., 2018), “P. sihuiensis” (Wu et al., 2014) and “P. wenzhouensis” (Zhang et al., 2021)], whose names have not been validly published. Because of their non-validly published status, new name combinations for these species are not proposed. However, in view of their reliable grouping with the Oleovorans clade, it is suggested that these species should also be recognized as members of the genus Ectopseudomonas with the names “E. indoloxydans,” “E. sediminis,” “E. sihuiensis” and “E. wenzhouensis,” respectively.

The Straminea clade is a strongly supported cluster encompassing seven Pseudomonas species (P. argentinensis, P. daroniae, P. dryadis, P. flavescens, P. punonensis, P. seleniipraecipitans, P. straminea) (Figure 1; Supplementary Figure S2). Species from this clade have also been found to group together in earlier studies (Hesse et al., 2018; Girard et al., 2021; Lalucat et al., 2022; Passarelli-Araujo et al., 2022). The members of this clade can be reliably distinguished from all other Pseudomonadaceae species by 12 novel CSIs identified in this study, which in most cases are exclusively shared by the species from this clade. Sequence information for one of these CSIs consisting of a three aa insertion in the protein Di-trans, poly-cis-decaprenylcistransferase is presented in Figure 3C. Detailed sequence information for this CSI and the 11 other CSIs specific for this clade are presented in Supplementary Figures S20–S31 and some of their characteristics are listed in Table 3. Based on the presented results showing the distinctness of this clade, we are proposing the transfer of species from this clade into Phytopseudomonas gen. nov.

The genus Stutzerimonas was recently described by Lalucat et al. (2022) by the transfer of several Pseudomonas species which branched distinctly in their phylogenetic tree. The clade labeled as Stutzerimonas in our phylogenetic tree (Figure 1) encompasses all 13 named Stutzerimonas species, whose genome sequences were available in the NCBI database at the time of analysis, as well as five non-validly published Pseudomonas species. Apart from their clustering in phylogenetic trees, there is no known reliable characteristic which is specific for the members of this genus. Our analyses have identified seven CSIs in different proteins, which in most cases are uniquely shared by all/most species from this clade. Sequence information for one of these CSIs is shown in Figure 4A. In this instance, a one aa insertion in a conserved region of the PAS domain-containing methyl-accepting chemotaxis protein is uniquely shared by all species from the Stutzerimonas clade. Detailed sequence information for this CSI and the six other CSIs specific for this clade/genus is provided in Supplementary Figures S32–S38 and some of their characteristics are summarized in Table 3. The identified CSIs provide reliable means for distinguishing Stutzerimonas species from all other Pseudomonadaceae species. Hence, we are emending the description of this genus to include these diagnostic characteristics.

Five species with non-validly published names [viz.“P. lopnurensis” (Mamtimin et al., 2021), “P. phenolilytica” (Kujur and Das, 2022), “P. oligotrophica” (Zhang et al., 2022), “P. saudiphocaensis” (Azhar et al., 2017) and “P. songnenensis” (Zhang et al., 2015)], also group reliably within the Stutzerimonas clade and share CSIs specific for this clade. These species should also be recognized as members of this genus with the names “S. lopnurensis,” “S. phenolilytica,” “S. oligotrophica,” “S. saudiphocaensis” and “S. songnenensis” respectively.

The Linyingensis clade consists of six Pseudomonas species viz., P. aromaticivorans, P. guangdongensis, P. linyingensis, P. oryzagri, “P. oryzae” and P. sagittaria, which form a strongly supported clade in our phylogenetic trees (Figure 1; Supplementary Figure S2). This clade is also denoted as g_Pseudomonas_K in the GTDB taxonomy (Parks et al., 2018). A specific evolutionary relationship among these species is supported by 15 CSIs (Table 3), which in most cases are uniquely shared by all species from this clade. In Figure 4B, we present one example of a CSI specific for this clade, where a five aa insertion in UDP-N-acetylmuramoyl-L-alanine--D-glutamate ligase protein is uniquely shared by all members of this clade. Detailed sequence information for this CSI and 14 other CSIs specific for this clade is presented in Supplementary Figures S39–S53. Based on these results, which robustly demarcate this species clade, we are proposing the transfer of these species into Geopseudomonas gen. nov.

The Resinovorans clade (Figure 1; Supplementary Figure S2), which is denoted as the taxon g_Pseudomonas_F in GTDB taxonomy (Parks et al., 2018), consists of six species viz. P. boanensis, P. furukawaii, P. lalkuanensis, P. otitidis, P. resinovorans and P. tohonis. Species from this clade also formed a distinct clade in earlier studies (Girard et al., 2021; Lalucat et al., 2022; Passarelli-Araujo et al., 2022). The members of this clade can be reliably distinguished from all other Pseudomonadaceae species by five identified CSIs, which in most cases are exclusively shared by all/most species from this clade. One example of a CSI specific for this clade is presented in Figure 5A, where in the Murein L, D-transpeptidase catalytic domain family protein, a two aa insertion is exclusively present in all species from the Resinovorans clade. Detailed sequence information for this CSI and four other identified CSIs, specific for this clade, is presented in Supplementary Figures S54–S58 and some of their characteristics are listed in Table 4. Based on these results, we are proposing the transfer of species from Resinovorans clade into Metapseudomonas gen. nov.

Oryzihabitans clade (denoted as the taxon g_Pseudomonas_B in GTDB taxonomy) consists of seven named Pseudomonas species viz. P. asuensis, P. duriflava, P. luteola, P. oryzihabitans, P. psychrotolerans, P. rhizoryzae and P. zeshuii, which form a strongly supported clade in our phylogenetic trees (Figure 1; Supplementary Figure S2). These species also formed a distinct clade in earlier phylogenetic studies (Hesse et al., 2018; Girard et al., 2021; Saati-Santamaría et al., 2021; Passarelli-Araujo et al., 2022). The best-studied species from this clade is P. luteola, which was originally a member of the genus Chryseomonas (Holmes et al., 1986). However, in 1997, based on 16S rRNA gene sequence similarity, this species was transferred into the genus Pseudomonas (Anzai et al., 1997). More recently, based on genomic studies, this species along with two other Pseudomonas species (P. asuensis and P. duriflava) were transferred into the genus Chryseomonas. It should be noted that C. luteola is a synonym of C. polytricha (Holmes et al., 1986), which is the type species of genus Chryseomonas (Parte et al., 2020). The genetic distinctness of the clade formed by these seven species is strongly supported by 11 novel identified CSIs which are uniquely shared by these species. One example of a CSIs specific for this clade is shown in Figure 5B. In this case, a one aa insertion in the protein cytochrome d ubiquinol oxidase subunit II is exclusively shared by all members of this clade. Detailed sequence information for this CSI and 10 other CSIs specific for this clade are presented in Supplementary Figures S59–S69 and some of their characteristics are listed in Table 4. In addition to the three species which are presently assigned to the genus Chryseomonas, four additional Pseudomonas species viz. P. oryzihabitans, P. psychrotolerans, P. rhizoryzae and P. zeshuii reliably group within this clade and share different CSIs specific for this genus. Hence, we are proposing new name combinations of these species to transfer them into the genus Chryseomonas.

The Thermotolerans clade includes the species P. carbonaria, P. cavernae, P. insulae and P. thermotolerans, which form a distinct clade in our phylogenomic trees (Figure 1; Supplementary Figure S2). Species from this clade also formed a distinct cluster in earlier studies (Girard et al., 2021; Lalucat et al., 2022). A specific evolutionary relationship among these species is strongly supported by five CSIs, which are exclusively shared by all members of this clade. One example of a CSI specific for this clade is shown in Figure 6A, where a six aa insertion in the TerB family tellurite resistance protein is exclusively found in all four species from this clade. Detailed sequence information for the five CSIs specific for this clade are presented in Supplementary Figures S70–S74 and some of their characteristics are listed in Table 4. Based on these results, we are proposing the transfer of species from this clade into Zestomonas gen. nov.

Pseudomonas flexibilis, formerly known as Serpens flexibilis (Hespell, 1977) was recently transferred into the genus Pseudomonas based on 16S rRNA similarity with P. pseudoalcaligenes (Shin et al., 2015). In our phylogenomic tree (Figure 1), this species branches separately from other Pseudomonas species and forms a distinct clade together with a newly described non-validly published species “Serpens gallinarum” (Gilroy et al., 2021) and another species P. tuomuerensis, which according to Shin et al. (2015) is a heterotypic synonym of P. flexibilis. This clade is identified as the taxon g_Pseudomonas_H in the GTDB taxonomy (Parks et al., 2018). A close and specific relationship of P. flexibilis (P. tuomuerensis) to “S. gallinarum” is independently supported by three CSIs identified in this study, which are exclusively shared by these species. One example of a CSI specific for this clade is shown in Figure 6B, where a one aa insertion in the protein GTP diphosphokinase is specifically shared by these three species. Detailed sequence information for this CSI and the two other CSIs specific for this clade is presented in Supplementary Figures S75–S77 and some of their characteristics are summarized in Table 4. Based on these results we are presenting an emended description of the genus Serpens with S. flexibilis as its type species.

The Fluvialis clade consists of the species P. fluvialis and P. pharmacofabricae, which formed a strongly supported clade in different phylogenetic trees (Figure 1; Supplementary Figure S2). Our analyses have identified eight CSIs in different proteins that are uniquely shared by these two species. Figure 7A depicts an example of a CSI, consisting of a seven aa deletion within a conserved region of an ATP binding protein, which is exclusively shared by these two species. Detailed sequence information for this and the six other CSIs specific for the Fluvalis clade is presented in Supplementary Figures S78–S85 and a summary of some of their sequence characteristics is presented in Table 5. Based on the results presented here, we are proposing the transfer of species from this clade into Caenipseudomonas gen. nov.

The genus Azotobacter was described by Beijerinck (1901) and its members are known to branch in between Pseudomonas species (Young and Park, 2007; Özen and Ussery, 2012; Lalucat et al., 2022). Four Azotobacter species whose genome sequences were analyzed in this study (viz. A. beijerinckii, A. chroococcum, A. salinestris, and A. vinelandii), formed a distinct clade branching in the proximity of Stutzeri and Linyingensis clades (Figure 1; Supplementary Figure S2). Similar branching of Azotobacter species has been reported in earlier work (Jun et al., 2016; Hesse et al., 2018; Lalucat et al., 2022). Our analyses have identified 10 CSIs which are exclusively found in all four Azotobacter species providing reliable means for the demarcation of this clade. Partial sequence information for one of the CSIs specific for this genus, found in the alginate export family protein, is shown in Figure 7B. Detailed sequence information for this CSI and nine other CSIs specific for this genus is provided in Supplementary Figures S86–S95, and some of their sequence characteristics are listed in Table 5.

Azomonas is another genus whose members branch in between Pseudomonas species (Figure 1; Supplementary Figure S2; Young and Park, 2007; Kennedy and Rudnick, 2015; Rudra and Gupta, 2021; Lalucat et al., 2022). The two Azomonas species included in our analyses (viz., A. agilis and A. macrocytogenes) formed a distinct cluster in our phylogenomic trees (Figure 1; Supplementary Figure S2). The distinctness of this clade is also supported by five CSIs identified in this work, which are exclusively shared by these two species. Sequence information for one of these CSIs, containing a five aa insertion within the protein succinate dehydrogenase flavoprotein, is shown in Figure 7C. Detailed sequence information for this CSI and the other four CSIs specific for this genus are provided in the Supplementary Figures S96–S100, and a summary of some of their sequence characteristics is listed in Table 5.

Aquipseudomonas (A.qui.pseu.do.mo’nas. L. fem. n. aqua, water; N.L. fem. n. Pseudomonas, a bacterial genus; N.L. fem. n. Aquipseudomonas, Pseudomonas-like species isolated from water).

Cells are Gram-stain negative, motile and rod shaped. The species are aerobic in respiration and have been isolated from soil and swimming pool water. Optimum temperature for growth ranges from 30 – 37°C with <2% (w/v) NaCl and pH range from 4–10. Genome sizes for the species vary from 4.3 Mb to 4.6 Mb and the GC content ranges from 63.3 to 65.5%. Of the species from this genus, the type species A. alcaligenes can degrade polycyclic aromatic hydrocarbons and has been proven useful for bioremediation of oil pollution, pesticide substances, and certain chemical substances. Species from this genus form a strongly supported clade in phylogenomic tree based on large datasets of concatenated proteins. Additionally, species from this genus can be reliably distinguished from all other Pseudomonadaceae genera based on six CSIs (Table 2) which are exclusively found in the species from this genus. New name combinations for the species that are part of this genus are provided in Table 7.

The type species of this genus is Aquipseudomonas alcaligenes.

Caenipseudomonas (Cae.ni.pseu.do.mo’nas. L. neut. n. caenum, mud; N.L. fem. n. Pseudomonas, a bacterial genus; N.L. fem. n. Caenipseudomonas, Pseudomonas-like organism(s) isolated from river sediments).

Cells are strictly aerobic, Gram-stain-negative, non-fluorescent and occur mostly as short rods. Cells are motile and contain a single polar flagellum. Chemoorganotrophic growth. Species have been isolated from river sediment, and wastewater sample from a pharmaceutical company. Growth occurs in the temperature range from 4-22°C with optimum growth occurring between 25-35^o C at pH between 7–8 in presence of 0–2% (w/v) NaCl concentration. Genome size range is from 3.3–3.4 Mb and the GC content is 62.6%. Species from this genus form a distinct lineage in phylogenomic trees based on large datasets of proteins, as well as in trees based on rpoD gene, or concatenated partial sequences for the 16S rDNA, gyrB, rpoB, and rpoD genes. In addition, species from this genus can be reliably distinguished based on eight exclusively shared CSIs listed in Table 5. The new name combinations for species from this genus are provided in Table 7.

The type species is Caenipseudomonas fluvialis.

Ectopseudomonas (Ec.to.pseu.do.mo’nas. Gr. prep. Ecto, outside; N.L. fem. n. Pseudomonas, a bacterial genus; N.L. fem. n. Ectopseudomonas, a genus outside of Pseudomonas).

Cells are Gram-stain negative, motile and rod shaped. Excepting E. chengduensis, all other species from this genus are motile due to the presence of a polar flagellum. Species have been isolated from diverse sources including sea water, soil, hot spring, compost, and lake sediments, etc. Chemoorganotrophic life cycle. Most species grow aerobically; however, some are indicated to be facultatively anerobic. Colonies are generally brownish yellow. Growth can occur from 4^o-42°C with optimum growth temperature between 30–37°C, with or without NaCl, in the pH range from 3.0–10.5 (optimum between pH 6–8). Genome sizes for known species vary from 4.5 Mb to 5.6 Mb and the GC content ranges from 62.2 to 65.0%. Of the species from this genus, E. mendocina can degrade toluene and it is indicated to cause opportunistic nosocomial infections. Members of this genus form a monophyletic clade in phylogenetic trees based on concatenated sequences of several large datasets of core genome proteins. Additionally, species from this genus also generally cluster together in phylogenetic trees based on rpoD gene, or concatenated partial sequences for the 16S rDNA, gyrB, rpoB, and rpoD genes. In addition of their distinct branching in phylogenetic trees, members of this genus can be reliably distinguished from other Pseudomonadaceae species based on five CSIs (Table 2) which in most cases are exclusively shared by the members of this genus. The new name combinations for species that are part of this genus are provided in Table 7.

The type species of this genus is Ectopseudomonas oleovorans.

Geopseudomonas (Ge.o.pseu.do.mo’nas. Gr. fem. n. gê, the Earth; N.L. fem. n. Pseudomonas, a bacterial genus; N.L. fem. n. Geopseudomonas, Pseudomonas like organisms isolated from soil).

Strictly aerobic to facultatively anaerobic, rod-shaped bacteria. Motile due to the presence of one or more polar or peritrichous flagella. Chemoorganotrophs, with cells exhibiting Gram-stain negative staining response. Cells generally do not produce fluorescent pigments. Members have been isolated from diverse sources including paddy soil, electroactive biofilm, herbicide applied wheat field and oil contaminated soil. Optimum growth occurs in the range of 30–37°C, between pH 7–8, in medium containing 1–2% NaCl (w/v). Genome lengths of the species vary from 3.2 to 4.7 Mb, and GC contents vary from 66.4 to 68.3%. Members of this genus form a monophyletic clade in phylogenetic tree based on concatenated sequences for several large datasets of proteins. Species from this genus also cluster together in phylogenetic trees based on rpoD gene, or concatenated partial sequences for the 16S rDNA, gyrB, ropB, and rpoD genes. In addition, the members of this genus can be reliably distinguished from all other Pseudomonadaceae genera by the 15 CSIs described in Table 3, which in most cases are exclusively shared by either all or most species from this genus. The new name combinations for species which are part of this genus are provided in Table 7.

The type species is Geopseudomonas sagittaria.

Metapseudomonas (Me.ta.pseu.do.mo’nas. Gr. adv. Meta, besides; N.L. fem. n. Pseudomonas, a bacterial genus; N.L. fem. n. Metapseudomonas, a genus beside Pseudomonas).

Species of this genus are Gram-negative, motile, aerobic and rod shaped. Chemoorganotrophic growth, cells do not produce fluorescent pigments. Members have been isolated from different sources such as clinical samples, soil or oil of wood mills and biphenyl contaminated soil. Optimum growth temperature is in the range of 30-37°C. Genome sizes for known species are in the range of 6.1 Mb to 6.8 Mb and GC content varies from 64.2 to 66.80%. Species from this genus form a strongly supported clade in phylogenomic trees based on large datasets of proteins. In addition, most of the species from this genus also cluster together in phylogenetic trees based on rpoD gene, or concatenated partial sequences for the 16S rDNA, gyrB, ropB, and rpoD genes. Importantly, the species from this genus can also be reliably distinguished from all other Pseudomonadaceae genera by the shared presence of five CSIs listed in Table 4. The new name combinations for the species of this genus are provided in Table 8.

The type species of this genus is Metapseudomonas resinovorans.

Phytopseudomonas (Phy.to.pseu.do.mo’nas. Gr. neut. n. phyton, plant; N.L. fem. n. Pseudomonas, a bacterial genus; N.L. fem. n. Phytopseudomonas, Pseudomonas-like species isolated from plants).

Cells are Gram-stain negative, motile due to the presence of a polar flagellum, aerobic, and rod shaped. Chemoorganotrophs. Most species have been isolated from different plant sources such as Quercus robur stem tissues, straw grass, rice paddy, walnut blight cankers etc. All species produce a diffusible fluorescent pigment. Optimum temperature for growth is between 25-30°C, with <4% (w/v) or without NaCl in the pH range from 6–8. Genome sizes for the species vary from 4.5 Mb to 5.9 Mb and the GC content ranges from 61.5 to 65.0%. Members of this genus form a monophyletic clade in phylogenetic trees based on concatenated sequences of several large datasets of core genome proteins. Additionally, species from this genus also generally cluster together in phylogenetic trees based on rpoD gene, or concatenated partial sequences for the 16S rDNA, gyrB, ropB, and rpoD genes. Additionally, members of this genus can be reliably distinguished from other Pseudomonadaceae genera based on the presence of 12 CSIs summarized in Table 3. which in most cases are exclusively present in the species from this genus. The new name combinations for species that are part of this genus are provided in Table 8.

The type species of this genus is Phytopseudomonas straminea.

Zestomonas (Zes.to.mo’nas. Gr. masc. Adj. zestos, hot, boiling; L. fem. n. monas, a unit, monad; N.L. fem. n. Zestomonas, a monad that can grow at high temperature).

Aerobic, motile rods exhibiting Gram-negative staining response. Chemoorganotrophs. Species have been cultivated from different sources such as cooking water, forest soil, charcoal, and cave sediment. Temperature range for growth for species from this genus differs considerably. While the optimum growth of the type species Zestomonas thermotolerans occurs at 47°C (growth range 25–56°C), other species from this genus grow optimally at 28–30°C. Genome length ranges from 3.8 to 5.5 Mb and the GC content varies from 64.5 to 66.8%. Members of this genus form a monophyletic clade in phylogenomic tree based on concatenated sequences for several large datasets of proteins. In addition, members of this genus can be reliably distinguished from other Pseudomonadaceae genera by their uniquely sharing five CSIs listed in Table 4. New name combinations for the species from this genus are provided in Table 8.

The type species is Zestomonas thermotolerans.

Azomonas (A.zo.mo.nas. N.L. pref. Azo-, pertaining to nitrogen; L. fem. n. monas, a unit, monad; N.L. fem. n. Azomonas, nitrogen monad).

Description of this genus is in large part based on that provided by Kennedy and Rudnick (2015) in the Bergey’s Manual of Systematics of Archaea and Bacteria. Cells are Gram-stain variable or sometimes Gram-stain negative depending on the culture age, aerobic, ellipsoidal to rod shaped. Species are motile with peritrichous or lophotrichous polar flagella. Cells may occur singly, in pairs, or in clumps. All species fix atmospheric nitrogen under aerobic conditions. Alternative nitrogenases containing vanadium (nitrogenase-2) or iron (nitrogenase-3) may only be synthesized in Mo-deficient media. Cultures can grow both aerobically and microaerobically. Chemoorganotrophic. Sugars, alcohols, and organic acids are used as carbon sources. Ammonium salts and sometimes nitrate (A. insignis only) are used as nitrogen sources; amino acids are not used. Water-soluble and fluorescent pigments are produced by nearly all strains. Species are catalase positive. The optimum pH for nitrogen fixation is close to neutrality, but certain strains can also fix nitrogen at a pH of 4.6–4.8. Species isolated from water or soil. The G + C content of DNA from known species varies from 52.0–58.6% and their genome size ranges from 3.3 to 4.1 MB. Species belonging to this genus form a distinct clade in phylogenomic trees based on concatenated sequences of large number of proteins and in the tree based on 16S rRNA gene sequences. In addition, members of this genus can be reliably distinguished from Azotobacter as well as all other Pseudomonadaceae genera based on their exclusive sharing five CSIs described in this work (Table 5).

Type species is Azomonas agilis (Beijerinck, 1901) Winogradsky, 1938 (Approved Lists 1980).

Azotobacter (A.zo.to.bac.ter. N.L. neut. n. azotum, nitrogen; N.L. masc. n. bacter, a rod; N.L. masc. n. Azotobacter, a nitrogen rod).

Description of this genus is in large part based on that provided by Kennedy et al. (2015) in the 2015 Bergey’s Manual of Systematics of Archaea and Bacteria. Cells range from straight rods with rounded ends to more ellipsoidal or coccoid. Motile with peritrichous flagella or nonmotile. Aerobic, having a strictly respiratory type of metabolism with oxygen as the terminal electron acceptor. Nitrogen is fixed under microaerobic conditions (2% oxygen), under full aerobiosis, or after adaptation in hyperbaric oxygen. N₂ fixation uses Mo-, V-, or Fe-containing nitrogenase enzymes, depending on the environmental metal supply. Water-soluble and water-insoluble pigments are produced by some strains. Growth is heterotrophic; sugars, alcohols, and salts of organic acids are used as carbon sources. Ammonium salts, nitrate, and urea are used as sources of fixed nitrogen. The pH range for growth is from 4.8 to 8.5, with optimum pH for diazotrophic growth between 7.0–7.5. Most isolates are from soil, but a few are from water. The GC content of the DNA varies from 65.5–67.5%. Genome size ranges from 4.9–5.4 Mb. Species belonging to this genus group together in phylogenetic trees based on 16S rRNA gene sequences, and in phylogenomic trees based on concatenated sequences of large number of proteins. In addition, members of this genus can be reliably distinguished from all other Pseudomonadaceae genera by 10 uniquely shared CSIs listed in Table 5.

Type species is Azotobacter chroococcum Beijerinck, 1901 (Approved Lists 1980).

Chryseomonas (Chry.se.o.mo’nas. Gr. masc. Adj. chryseos, golden; L. fem. n. monas, a unit, monad; N.L. fem. n. Chryseomonas, a yellow unit).

The description of this genus is partially based on that given by Holmes et al. (1986) for the type species (C. polytricha) of this genus. The cells are rod-shaped, Gram-negative, aerobic, and exhibit chemoorganotrophic growth. Except for C. duriflava (and its synonym C. zeshuii), which do not exhibit motility, cells from the other species are motile by either a single or several polar or trichous flagella. Known species have been isolated from diverse sources including rice seeds and paddy, desert soil, herbicide-contaminated soil, grass rhizosphere, clinical specimens, and medical clinic for small animals. C. oryzihabitans has been reported as pathogenic to plants and animals. Some species (C. luteola) can reduce nitrate. Growth can occur in the temperature range from 4–42°C with optimum growth occurring between 30 to 37°C at pH 7.0 (pH range 6–8) in medium supplemented with 1–2% (w/v) NaCl. The cells are catalase positive but oxidase negative. The GC content of species varies from 53.6 to 66.2% and their genome lengths range from 4.3 to 5.4 Mb. Species from this genus form a distinct clade in the phylogenomic trees based on a large number of proteins. Additionally, these species also cluster together in phylogenetic trees based on rpoD gene, or concatenated partial sequences for the 16S rDNA, gyrB, ropB, and rpoD genes. Apart from their grouping together in phylogenetic trees, species from this genus can be reliably distinguished from all other Pseudomonadaceae genera by their 11 CSIs listed in Table 4, which in most cases are exclusively present in the species from this genus. New name combinations for four Pseudomonas species, which are transferred to this genus, are provided in Table 8.

Type species of this genus is Chryseomonas polytrichia (Holmes et al., 1986).

Description of this genus is modified from that given by Hespell (1977). Gram-negative, aerobic, rod-shaped, non-spore forming, bacterial cells. Cells from the type species, S. flexibilis, are very flexible, and motile due to containing a flagellum, and exhibit serpentine-like movement in agar gels. Metabolism is respiratory, and molecular oxygen serves as the terminal electron acceptor. S. flexibilis mainly uses lactate as the energy and carbon source. Catalase and oxidase are produced. Temperature range for optimal growth is from 28 to 37°C. The G + C content of DNA ranges from 61.0–65.8 mol% and genome size varies from 3.8–3.9 Mb. Species from this genus form a monophyletic clade in the phylogenetic tree based on large dataset of proteins. The type species also forms a distinct lineage in phylogenetic trees based on rpoD gene, or concatenated partial sequences for the 16S rDNA, gyrB, ropB, and rpoD genes. Additionally, species from this genus can be reliably distinguished from other Pseudomonadaceae genera by the presence of three exclusively shared CSIs (Table 4). New name combinations for the two species which are part of this genus are provided in Table 8.

Type species of this genus is Serpens flexibilis Hespell, 1977 (Approved lists).

Stutzerimonas (Stut.ze.ri.mo’nas. L. fem. n. monas, a unit, monad; N.L. fem. n. Stutzerimonas, monad of Stutzer, named in honor of Albert Stutzer, who in 1895 described the bacterium today known).

The description of this genus, especially in terms of its morphological, chemotaxonomic and growth characteristics, remains the same as provided by Lalucat et al. (2022). In addition to the genomic characteristics described by Lalucat et al. (2022), members of this genus can be reliably distinguished from other Pseudomonadaceae genera by seven novel CSIs identified in this study (listed in Table 3), which in most cases are exclusively found in the species from this genus. New name combination for P. marianensis (Table 8) is based on its branching in the 16S rRNA gene tree (Yang et al., 2022).

The type species is Stutzerimonas stutzeri (Lehmann and Neumann 1896) Lalucat et al. 2022.