Supplementary Materials for
Public good exploitation in natural bacterioplankton communities
Shaul Pollak, Matti Gralka, Yuya Sato, Julia Schwartzman, Lu Lu, Otto X. Cordero*
*Corresponding author. Email: [email protected] Published 28 July 2021, Sci. Adv. 7, eabi4717 (2021)
DOI: 10.1126/sciadv.abi4717
The PDF file includes:
Figs. S1 to S14 Tables S1 to S4
Legends for data S1 to S5
Other Supplementary Material for this manuscript includes the following:
Data S1 to S5
Fig. S1. Distribution of KEGG annotations in chitinase linked genes. Annotations were inferred using eggnog-mapper. Annotations are weighted in each gene, so that the total number of annotations in each gene is 1. The total number of annotations reported here is the sum of all weighted annotations across all genes. Mobile genes are genes that had one of the following terms in their description: phage, tail, capsid, transposon, transposase, conjugation, or insertion element. Unknown function genes either had no match to an eggnog- mapper entry, had no KEGG pathway annotation, or contained the term “unknown function” in the description field.
Fig. S2. enrichment of KEGG carbon metabolic pathways in chitinase-linked genes. Enrichment is calculated relative the fraction of genes with the specified annotation in full gene catalogue (I.e, if 10% of all genes are annotated as belonging to pathway X, but 20% of chitinase linked genes belong to pathway X, the enrichment is 2-fold). Bars are colored black if the hypergeometric p-value is lower than 0.05.
E nr ic hm en t fa ct or
Glycolysis / GluconeogenesisTCA cycle Pentose Phosphate Pathway Pentose and glucuronate interconversions Fructose and Mannose Metabolism
Galactose Metabolism
Ascorbate and aldarate metabolism
Starch and Sucrose Metabolism Pyruvate Metabolism Glyoxylate and dicarboxylate metabolism Propanoate Metabolism Butanoate Metabolism
C5-Branched dibasic acid metabolism Inositol phosphate metabolism Amino sugar and nucleotide sugar metabolism
Hypergeometric p-value p-value < 0.05 p-value ≥ 0.05
Fig. S3. Density of conditional gene loss probabilities, given that the other gene category is retained. The probability was calculated for each gene family across all edges of the relevant sub-tree over which the gene family exists. Unlike Fig. 1D, here the density is plotted with a logarithmic y axis and linear x axis, allowing the comparison of the densities at 0 conditional probability (0 x-axis values).
0.01 0.10 1.00 10.00
0. 00
0. 25
0. 50
0. 75
1. 00
D en si ty
P(chitinase lost | linked gene retained)
P(linked gene lost | chitinase gene retained)
Fig. S4. Chitinase linked genes are better than random sets of genes at predicting genomic chitinase copy number. For each phylum/class the ability of our elastic-net linear regression to predict chitinase copy number, in terms of the R2 of the regression, was compared to 1000 models trained with the same parameters on an equal number of randomly selected genes. The distribution of R2 values of these random gene-based elastic-net linear models is plotted for each phylum/class, where the indicated p-value is the proportion of cases that a random gene elastic-net model had a higher R2 than the chitinase linked genes-based elastic-net linear model. In cases where the p-values were insignificant, the successful random sets of genes either contained a large amount of co-evolving genes or contained many genes that were negatively linked to chitinases (data not shown). These negatively linked genes also contain valuable information about chitinase copy-number, but their functional significance in the context of chitin degradation is unclear.
0 30 60 90
0.84 0.88 0.92 0.96
R^2
count
Actinobacteriota, pvalue = 0.002
0 50 100 150
0.76 0.80 0.84 0.88 0.92
R^2
count
all, pvalue = 0
0 25 50 75
0.1 0.2 0.3 0.4 0.5 0.6
R^2
count
Alphaproteobacteria, pvalue = 0.002
0 50 100
0.2 0.3 0.4 0.5 0.6 0.7
R^2
count
Bacteroidota, pvalue = 0.002
0 50 100 150 200 250
0.0 0.2 0.4 0.6
R^2
count
Cyanobacteria, pvalue = 1
0 30 60 90
0.80 0.84 0.88 0.92
R^2
count
Firmicutes, pvalue = 0.003
0 50 100
0.75 0.80 0.85
R^2
count
Gammaproteobacteria, pvalue = 0
0 30 60 90
0.7 0.8 0.9 1.0
R^2
count
other, pvalue = 0.422
Fig. S5. Sensitivity of cluster presence threshold. The threshold represents the minimal frequency of genes in each cluster found in a genome in order for the cluster to be considered present in that genome. For each threshold, the model was trained on publicly available genomes, as in the main text, and accuracy was assessed on our genome collection.
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.0
0.1 0.2 0.3 0.4 0.5 0.6 0.7
P ro p o rt io n o f p h e n o ty p e s p re d ic te d c o rr e ct ly
Cluster presence threshold
Fig. S6. Predicted ecological classes of 63 marine isolates (out of which 57 were phenotyped). For each bacterial order in our collection (as determined by GTDB-tk), we predicted the expected number of chitinases (Echi) based on the chitinase linked gene content of the isolate’s genome, and compared it to the observed number of chitinases. All ecological roles were found in multiple orders and families, and for some orders contained multiple ecological roles, suggesting that ecological roles evolve faster than, and are unlinked from phylogeny at the order level.
E n te ro b a c te ra le s, A lte ro m o n a d a c e a e E n te ro b a c te ra le s, P sy c h ro b ia c e a e E n te ro b a c te ra le s, P sy c h ro m o n a d a c e a e E n te ro b a c te ra le s, V ib ri o n a c e a e F la vo b a c te ri a le s, F la vo b a c te ri a c e a e P s e u d o m o n a d a le s, H a lo m o n a d a c e a e P s e u d o m o n a d a le s, N itr in c o la c e a e P s e u d o m o n a d a le s, O le ip h ila c e a e P s e u d o m o n a d a le s, S a c c h a ro s p ir ill a c e a e R h o d o b a c te ra le s, R h o d o b a c te ra c e a e 0
5 10 15
# o f is o la te s
degrader
exploiter
scavenger
Fig. S7. Accurate quantification of bacterial cell counts on chitin using SYBR-gold. (A) fluorescence of individual strains as a function of cell density (CFU/ml). Cells were grown in marine broth 2216 to saturation and were then serially diluted and stained with SYBR-gold for 15 minutes, after which their fluorescence was determined. (B) P-values of the interaction term between strain identity and cell concentration for the regression lines of the data presented in panel A. No interaction term is significant (all p-values > 0.1), indicating that cell concentration influences fluorescence in a similar way across all tested strains. (C) Strain vibrio1A01 was stained with SYBR-gold in the presence of various concentrations of colloidal chitin. Particle dilutions are relative to the standard concentration used for assaying growth, which is 2 [g/l]. Cell concentrations are indicated as the dilution factor relative to the saturation cell density, which is
~109 [cells/ml].The shaded region indicates the dilution factor where particle fluorescence overwhelmed cellular DNA fluorescence. The dashed line indicates our detection threshold, which was used in Figure 2A to define growth.
10 100 1000 10000
Fluorescence [a.u]
1e+05 1e+06
1e+07 1e+08
1e+09 CFU/ml
genotype 1A01 6C06 6D02 6D03 I2M19 I2R16
A
0.00 0.25 0.50 0.75
p-value
6C06 6D02
6D03 I2M19
I2R16 Strain B
Noise
300 1000 3000 10000 30000
Fluorescence [a.u]
0.001 0.010
0.100 1.000 cell dilution
particle dilution -3 -2 -1 0
C
Detection limit
Fig. S8. Confusion matrix for the two models. For each model, the underlying data, which is the same data as presented in Fig. 2A, is presented on the left, where the points are colored according to the predictions of the indicated model. The confusion matrix that compares the predicted labels to observed labels is presented on the right. The major differences between the models are in their ability to identify exploiters.
n = 6
n = 1 n = 4
n = 11
n = 4 n = 4 n = 3
n = 24 scavenger
exploiter degrader
degrader exploiter
scavenger
Actual
Predicted
Chitinase-linked gene content
n = 6
n = 1 n = 4
n = 2
n = 13 n = 3 n = 1
n = 27 scavenger
exploiter degrader
degrader exploiter
scavenger
Actual
Predicted
Genome-scale metabolic model
n = 0 n = 0 1
10 100 1000 10000
1 10 100 1000 10000
degrader
exploiter
scavenger
1 10 100 1000 10000
1 10 100 1000 10000
degrader
exploiter
scavenger
Fluorescence on GlcNAc [a.u]
Fluorescence on GlcNAc [a.u]
Fluorescence on Chitin [a.u]Fluorescence on Chitin [a.u]
Fig. S9. Genome-scale metabolic models perform worse in understudied taxa at positive prediction tasks (predicting Degraders and Exploiters), compared to our (nearly) annotation free coevolving gene model. Each panel depicts a different taxonomic order.
The number isolates belonging to each ecological class according to their measured phenotype is depicted by full bars. The number of isolates correctly predicted by each method is depicted by hollow bars.
Chitinase-linked gene content
Pseudomonadales,Saccharospirillaceae Rhodobacterales,Rhodobacteraceae
Flavobacteriales,Flavobacteriaceae Pseudomonadales,Halomonadaceae Pseudomonadales,Nitrincolaceae Pseudomonadales,Oleiphilaceae Enterobacterales,Alteromonadaceae Enterobacterales,Psychrobiaceae Enterobacterales,Psychromonadaceae Enterobacterales,Vibrionaceae
degrader exploiter scavenger degrader exploiter scavenger
degrader exploiter scavenger degrader exploiter scavenger degrader exploiter scavenger degrader exploiter scavenger degrader exploiter scavenger degrader exploiter scavenger degrader exploiter scavenger degrader exploiter scavenger
0 1 2 3
0 1 2 3 0.00
0.25 0.50 0.75 1.00
0.0 0.5 1.0 1.5 2.0 0.00
0.25 0.50 0.75 1.00
0.00 0.25 0.50 0.75 1.00
0 5 10 0
2 4 6
0 2 4 6
0.00 0.25 0.50 0.75 1.00
Phenotype
# of isolates Total number
of isolates in each class
Genome-scale metabolic model
Fig. S10. Dynamics of an exploiter–degrader co-culture. (A) Illustration of the flat-chitin experimental setup used to measure co- culture dynamics. Strains are marked by fluorescent proteins expressed from a plasmid encoded constitutively expressed construct (methods) (B) Degrader-Exploiter co-culture reveals the consequence of exploitation among co-inhabiting environmental isolates. The Y- axis displays the fluorescence intensity, which is highly correlated by OD (Supplementary Figure S9). Grey lines represent mono-culture readings, while colored lines represent readings from co-cultures. Shaded regions represent standard deviations from the mean from three replicates. Vertical lines represent the earliest time where exploiter growth was observed when co-cultured with a degrader.
In co-culture, exploiter grows during degrader's rapid growth phase
Degrader Exploiter
RFP GFP
Chitinase GlcNAc
Flat Chitin sheet
RFP GFP
A B
Experimental setup to measure co-culture dynamics on chitin
Fluorescence [A.U]
Time [hours]
0 5000 10000 15000 20000
0 25 50 75
Degrader in mono- culture
Degrader in co-culture with exploiter
0 500 1000 1500 2000
0 25 50 75
Exploiter in mono-culture Exploiter in co-culture with exploiter Time to measurable exploiter growth
Fig. S11. Plasmid encoded fluorescence correlates with OD. (A,C) Fluorescence / Optical density (600nm, OD) as a function of time for strain vibrio1A01-DsRed (A) and strain alteroA3R04-eGFP (C). Red represents DsRed fluorescence, green represents eGFP fluorescence, and black represents OD. Experiments were conducted in triplicate, and the lines present the mean, with the shaded region representing the standard deviation. The DsRed reporter has a higher signal-to-noise compared with the eGFP reporter, resulting in higher background subtracted values, as well as the ability to detect an increase in fluorescence before an increase in OD. 1A01-DsRed
measurements were performed with the chitin sheet as a sole carbon source, as in Fig. 2. alteroA3R04 is unable to grow in monoculture on the chitin sheet, so measurements were performed using the same protocol, but with 20 [mM] N-Acetylglucosamine added, resulting in faster dynamics. (B,D) correlation between fluorescence and optical density for vibrio1A01-DsRed (B) and alteroA3R04 (D). Values are taken from A and D respectively.
0 5000 10000 15000 20000
Fluorescence
0 2 4
OD
0 25 50 75
Time [h]
A
0.2 0.4 0.6
Optical density [600nm]
0 5000 10000 15000 20000
RFP fluorescence [a.u]
B
500 1000 1500 2000 2500
Fluorescence
0.2 0.4 0.6
OD
0 5 10 15 20 25
Time [h]
C
0.0 0.1 0.2 0.3 0.4 0.5
Optical density [600nm]
500 1000 1500 2000
GFP fluorescence [a.u]
D
r = 0.98 r = 0.98
Measured on chitin sheet Measured on GlcNAc monomers
Fig. S12. Using plasmid-based fluorescence to measure co-culture dynamics on chitin sheets. (A,C) GFP and YFP fluorescence as a function of time for vibrio1A01-DsRed (A), and alteroA3R04-eGFO (C). Data is reproduced from experiments presented in Fig. S7, with the measurements of the additional fluorescent channel presented. Red lines represent DsRed fluorescence, Green lines represent eGFP fluorescence, and shaded regions represent standard deviations around the mean, taken from at least 2 indipendent replicates. (B,D) Correlation between DsRed and eGFP fluorescence for vibrio1A01-DsRed (B), and alteroA3R04-eGFP (D). The fit in B was used to correct eGFP values in co-cultures presented in panels E-G and in Fig. 2 in the main text. (E-G) co-culture dynamics of vibrio1A01- DsRed and alteroA3R04-eGFP when grown with a flat chitin sheet as the sole carbon source. To show the effect of the initial ratio of the two strains on the outcome of the co-culture, values were normalized such that the minimal DsRed/eGFP fluorescence value across all experiments was set to 0 and the maximal value across all experiments was set to 1. The data in F is used in Fig. 2 in the main text.
0 5000 10000 15000 20000
0 25 50 75
time [h]
fluorescence [a.u]
1A01-RFP A
750 1000 1250 1500 1750 2000
5000 10000 15000 20000
RFP fluorescence
GFP fluorescence
1A01-RFP B
0.0 0.2 0.4 0.6
0 25 50 75
time [h]
normalized fluorescence
A3R04_10%_1A01_90%
C
0.0 0.2 0.4 0.6
0 25 50 75
time [h]
normalized fluorescence
A3R04_50%_1A01_50%
D
0.0 0.2 0.4 0.6
0 25 50 75
time [h]
normalized fluorescence
A3R04_90%_1A01_10%
E
Fig. S13. Simple consumer-resource model captures co-culture yield patterns. (A) community and resource dynamics for 3 exploiter advantage ratios. The exploiter advantage ratio is defined as γexploiter / γdegrader. The model parameters were set to α = 2*10-4, γdegrader = 10-2, λ
= 10-2. Initial conditions were Chitin0 = 103, GlcNAc0 = 0, degrader0 = 10, and exploiter0 = 10. The model was simulated for 150 time steps. (B) maximal yield of simulated co-cultures as a function of the exploiter advantage ratio. The axes show the ratio between the yield in co-culture and the yield in monoculture for the degrader (x-axis) and exploiter (y-axis). The exploiter monoculture was assumed to go through 1 doubling, but this choice only modulates the numbers on the y-axis, and does not change the qualitative outcome of the figure.
The axes directly correspond to the axes in Figure 2C of the main text.
0 50 100 150
0 250 500 750 1000
0 50 100 150
0 2 4 6 8
0 50 100 150
0 100 200 300
0 50 100 150
0 100 200 300
0 50 100 150
200 400 600 800 1000
0 50 100 150
01 23 45 6
0 50 100 150
1020 3040 5060 70
0 50 100 150
0 100 200 300
0 50 100 150
500 600 700 800 900 1000
0 50 100 150
0 1 2 3 4 5
0 50 100 150
10 15 20 25 30
0 50 100 150
50 100 150 200 250
Time
Chitin GlcNAc Degrader Exploiter
advantage ratio
1
1.5
2
0.2 0.4 0.6 0.8 1.0
14 16 18 20
1.0 1.2 1.4 1.6 1.8 2.0
advantage ratio
Degrader yield in co-culture Degrader yield in mono-culture Exploiter yield in co-culture Exploiter yield in mono-culture
A
B
Fig. S14. Dynamics of strains belonging to different ecological classes in a highly complex natural seawater community. (A) dynamics of individual strains. Raw relative abundances were calculated as described in the methods section. Relative frequencies below the threshold value of 10-4 were set to that value for the purpose of this plot. The GTDB-Tk determined taxonomic family of each isolate is depicted on the right-hand side of the heatmap. (B) Mean relative abundance of each ecological class, averaged over all isolates belonging to that class, as a function of time.
A3R16 D2R184B034A104F10 A3M17B2R223B05 B2M13 G2R14C1R065C01 B3M02 C3M10F3R11 G3M19 B2M17D2R05A3R12C3R196B07 C2M11B3R186D02 C1M14 D2M026C063D051A016D03
0 50 100 150 200
3e-04 1e-03 3e-03 1e-02 3e-02
Mean relative abundance
0 50 100
Time [h]
degrader exploiter scavenger
-3 -2 -1 Relative -4
abundance
log10 Tax.
Fam.
Alteromonadac eae Flav obac teriaceae Halomonadac eae Nitrinc olac eae Oleiphilac eae Psychrobiaceae Psyc hromonadaceae Rhodobac terac eae Vibrionac eae GTDB-Tk taxonomic family
Time [h]
A B
DegraderExploiterScavenger
Table S1. Number of genomes analyzed in each phylum or class, and the index by which the phylum is indicated in Fig. 1A.
Phylum or class # of genomes number in figure 1A
Gammaproteobacteria 3550 1
Firmicutes 2194 4
Actinobacteriota 928 6
Alphaproteobacteria 711 2
other 492 none (grey edges)
Bacteroidota 467 7
Campylobacterota 295 3
Cyanobacteria 115 5
Table S2. Elastic-net regression summary statistics.
For each phylum, the number of genomes analyzed, number of input features (clusters of chitinase linked genes that are found in at least one genome in that phylum or class), number of variables used by the elastic-net regression model, mean-square error of the model, and 5-fold cross-validated R
2. Values were obtained from the output of the cv.glmnet function in the glmnet R package.
phylum n nvar nvar_used mse R
2all 8752 1905 910 0.26 0.91
Gammaproteobacteria 3550 882 360 0.34 0.87
Firmicutes 2194 565 225 0.20 0.91
Actinobacteriota 928 509 181 0.29 0.96
Alphaproteobacteria 711 194 42 0.06 0.57
other 492 254 56 0.25 0.90
Bacteroidota 467 137 103 0.34 0.73
Campylobacterota 295 19 1 0.02 0.06
Cyanobacteria 115 38 0 0.27 0.00
Table S3. Isolate phenotypes and model predictions.
The data underlying Fig 2A and Supplementary Figure S6 is given for each isolate in the columns labeled glcnac and chitin. The numbers represent the mean maximal fluorescence on each carbon source during 96 h of growth in monoculture. The phenotype is classification is derived from these measurements, where growth is considered as showing a fluorescence value > 1000. Degraders grow on both carbon sources, exploiters only on GlcNAc, and scavengers on neither. The predictions of both our co-evolution based model, and a CarveMe derived genome-scale metabolic model are indicated for each isolate. GTDB-tk derived phylogeny down to the genus level is also provided for each isolate
strain Phylum Class Order Family Genus GlcNAc chitin phenotype
Chitinase -linked
gene content
Genome- scale metaboli
c model 1A01 Proteobacteria Gammaproteobacteria Enterobacterales Vibrionaceae Vibrio 31108.2 11766.5 degrader degrader degrader 3B05 Proteobacteria Gammaproteobacteria Pseudomonadales Nitrincolaceae Neptunomonas 0.0 0.0 scavenger scavenger scavenger 3D05 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Pseudoalteromonas 23488.7 2420.3 degrader degrader degrader 4A10 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Epibacterium 0.0 0.0 scavenger scavenger scavenger 4B03 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Alteromonas 0.0 126.0 scavenger scavenger scavenger 4F10 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Pseudopelagicola 0.0 0.0 scavenger scavenger scavenger 5C01 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae OMPR01 230.0 0.0 scavenger scavenger scavenger 6B07 Bacteroidota Bacteroidia Flavobacteriales Flavobacteriaceae Maribacter 10437.7 0.0 exploiter exploiter scavenger 6C06 Proteobacteria Gammaproteobacteria Enterobacterales Psychromonadaceae Psychromonas 6232.2 7542.3 degrader degrader degrader 6D02 Proteobacteria Gammaproteobacteria Enterobacterales Psychrobiaceae Psychrobium 7233.9 1263.1 degrader exploiter exploiter 6D03 Proteobacteria Gammaproteobacteria Enterobacterales Vibrionaceae Vibrio 19031.7 19044.6 degrader degrader degrader A2M03 Bacteroidota Bacteroidia Flavobacteriales Flavobacteriaceae Zobellia 739.0 117.4 scavenger degrader scavenger A3M17 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Ruegeria 204.5 7.2 scavenger scavenger scavenger A3R04 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Alteromonas_B 23720.0 201.6 exploiter exploiter scavenger A3R06 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Tritonibacter 8193.3 0.0 exploiter scavenger scavenger A3R12 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Paraglaciecola 0.0 0.0 scavenger scavenger scavenger A3R16 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Paraglaciecola 0.0 0.0 scavenger scavenger exploiter B2M06 Bacteroidota Bacteroidia Flavobacteriales Flavobacteriaceae Cellulophaga 7946.9 0.0 exploiter exploiter scavenger B2M13 Proteobacteria Gammaproteobacteria Pseudomonadales Halomonadaceae Cobetia 2.1 8.8 scavenger scavenger scavenger B2M17 Bacteroidota Bacteroidia Flavobacteriales Flavobacteriaceae Polaribacter 0.0 156.2 scavenger exploiter scavenger B2R09 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Roseovarius 0.0 0.0 scavenger scavenger scavenger B2R22 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Octadecabacter 0.0 0.0 scavenger scavenger scavenger B3M02 Proteobacteria Gammaproteobacteria Enterobacterales Psychromonadaceae Psychromonas 23072.7 154.7 exploiter scavenger degrader B3M08 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Lentibacter 5860.0 0.0 exploiter scavenger scavenger B3R10 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Psychrosphaera 8867.0 11.9 exploiter exploiter scavenger B3R18 Bacteroidota Bacteroidia Flavobacteriales Flavobacteriaceae Zobellia 2898.8 0.0 exploiter exploiter scavenger C1M14 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Alteromonas_B 25228.0 153.0 exploiter degrader scavenger C1R06 Proteobacteria Gammaproteobacteria Pseudomonadales Nitrincolaceae Amphritea 0.0 4.4 scavenger exploiter scavenger C2M11 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Colwellia 7836.2 213.2 exploiter exploiter exploiter C2R02 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Pseudoalteromonas 0.0 0.0 scavenger degrader degrader C2R09 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Paracoccus 16.7 32.1 scavenger scavenger exploiter C3M06 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Salipiger 69.6 20.8 scavenger scavenger scavenger C3M10 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae OMPR01 10.4 0.0 scavenger scavenger scavenger C3R12 Proteobacteria Gammaproteobacteria Enterobacterales Vibrionaceae Vibrio 32589.0 113.7 exploiter degrader degrader C3R17 Bacteroidota Bacteroidia Flavobacteriales Flavobacteriaceae Zobellia 584.0 56.2 scavenger exploiter scavenger C3R19 Bacteroidota Bacteroidia Flavobacteriales Flavobacteriaceae Maribacter 2703.8 60.7 exploiter exploiter scavenger D2M02 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Colwellia 18222.9 4308.5 degrader degrader degrader D2M19 Proteobacteria Gammaproteobacteria Pseudomonadales Oleiphilaceae Marinobacter 0.0 34.8 scavenger scavenger scavenger D2R04 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Pacificibacter 0.0 0.0 scavenger scavenger scavenger D2R05 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Aliiglaciecola 5693.0 81.5 exploiter exploiter exploiter D2R18 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Yoonia 15614.3 0.0 exploiter scavenger scavenger
D3R19 Bacteroidota Bacteroidia Flavobacteriales Flavobacteriaceae Winogradskyella 34.0 132.1 scavenger scavenger scavenger E3M18 Bacteroidota Bacteroidia Flavobacteriales Flavobacteriaceae Arenibacter 5391.9 0.0 exploiter exploiter scavenger E3R01 Bacteroidota Bacteroidia Flavobacteriales Flavobacteriaceae Tenacibaculum 0.0 0.0 scavenger degrader scavenger E3R09 Bacteroidota Bacteroidia Flavobacteriales Flavobacteriaceae Winogradskyella 0.0 0.0 scavenger exploiter scavenger E3R11 Proteobacteria Gammaproteobacteria Enterobacterales Vibrionaceae Vibrio 21435.4 0.0 exploiter degrader degrader F2R14 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Pacificibacter 0.0 0.0 scavenger scavenger scavenger F3M07 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Psychrosphaera 0.0 115.0 scavenger scavenger scavenger F3R08 Proteobacteria Gammaproteobacteria Pseudomonadales Oleiphilaceae Marinobacter 69.8 154.3 scavenger scavenger scavenger F3R11 Proteobacteria Gammaproteobacteria Pseudomonadales Oleiphilaceae Marinobacter 7.7 0.0 scavenger scavenger scavenger G2M07 Proteobacteria Gammaproteobacteria Pseudomonadales Saccharospirillaceae Reinekea 502.2 0.0 scavenger scavenger exploiter G2R10 Proteobacteria Gammaproteobacteria Enterobacterales Vibrionaceae Vibrio 32814.7 6053.0 degrader degrader degrader G2R14 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Pseudopelagicola 0.0 0.0 scavenger scavenger scavenger G3M19 Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Celeribacter 0.0 0.0 scavenger scavenger scavenger I2M19 Bacteroidota Bacteroidia Flavobacteriales Flavobacteriaceae Tamlana_A 6087.7 0.0 exploiter exploiter scavenger I2R16 Proteobacteria Gammaproteobacteria Enterobacterales Alteromonadaceae Psychrosphaera 7157.1 0.0 exploiter exploiter scavenger I3M07 Proteobacteria Gammaproteobacteria Enterobacterales Vibrionaceae Vibrio 33343.3 88.8 exploiter degrader degrader
