Data from: Order among chaos: high throughput MYCroplanters can distinguish interacting drivers of host infection in a highly stochastic system
Data files
Jan 15, 2025 version files 624.24 MB
-
01_Analysispath_compratio.zip
141.21 MB
-
02_Analysispath_fluoranalysis.zip
26.90 MB
-
03_Analysispath_priorityeffects.zip
32.15 MB
-
04_Analysispath_priorityplates.zip
4 KB
-
05_SupplementalData.zip
167.86 KB
-
Healthscore_data.zip
421.49 MB
-
MYCROPLANTER_3DPRINTING.zip
2.29 MB
-
README.md
34.90 KB
Abstract
The likelihood that a host will be susceptible to infection is influenced by the interaction of diverse biotic and abiotic factors. As a result, substantial experimental replication and scalability are required to identify the contributions of and interactions between the host, and the environment, and biotic factors such as the microbiome. For example, pathogen infection success is known to vary by host genotype, microbiota strain identity and dose, and pathogen dose. Elucidating the interactions between these factors in vivo has been challenging because testing combinations of these variables quickly becomes experimentally intractable. Here, we describe a novel high throughput plant growth system (MYCroplanters) to test how multiple host, microbiota, and pathogen variables predict host health. Using an Arabidopsis-Pseudomonas host-microbiota-pathogen model, we found that host genotype and bacterial strain order of arrival predict host susceptibility to infection, but pathogen and microbiota dose can overwhelm these effects. Host susceptibility to infection is therefore driven by complex interactions between multiple factors that can both mask and compensate for each other. However, regardless of host or inoculation conditions, the ratio of pathogen to microbiota emerged as a consistent predictor of disease. Our results demonstrate that high-throughput tools like MYCroplanters can isolate interacting drivers host susceptibility to disease. Increasing the scale at which we can screen drivers of disease outcomes, such as microbiome community structure, will facilitate both disease predictions as well as engineering solutions for medicine and agricultural applications.
README: Data from: Order among chaos: high throughput MYCroplanters can distinguish interacting drivers of host infection in a highly stochastic system
https://doi.org/10.5061/dryad.w9ghx3fxd
This dataset is the complete collection of raw data for the manuscript, "Order among chaos: high throughput MYCroplanters can distinguish interacting drivers of host infection in a highly stochastic system" by Chen et al 2025 PLOS Pathogens.
Description of the data and file structure
Unless otherwise stated, NA means data are not applicable and a empty cell means data was not collected. Experiments were named with the naming convention: 'YYYY-MM-DD_[short description of experiment]', where the date is the day plants were inoculated.
01_Analysispath_compratio
This includes data and metadata from MYCroplanter experiments used to determine the effect of inoculation ratio on plant health outcome. This data were used in Figures 1, 2, and 3 from the main manuscript and Supplementary Figures S1-S9
00_raw_data/dat/abs_data/
- abs_to_od_data.csv : Data that relates absorbance of 100uL culture at 600nm to OD600 (1cm cuvette). Column descriptions are as follows:
- ‘isolate’ : Which Pseudomonas isolate was measured
- ‘Concentration’ : Dilution of overnight culture, where 1 = no dilution and 0 = pure LB
- ‘ABS600’ : Absorbance reading in a 96-well plate of 100uL of culture at 600nm
- ‘OD600’ : Absorbance reading of 1mL of culture in a standard 1cm cuvette at 600nm. NA = Not available
- ODs_beforeafter_longformat.csv : Absorbance readings of bacterial cultures at 600nm after 5, 6, and 7 days of plant-microbe co-coculture in the MYCroplanter system. Column descriptions are as follows:
- ‘plate’ : Plate name; corresponds to directory name of pixel_data
- ‘row’ : Row in 96-well plate
- ‘col’ : Column in 96-well plate
- ‘od_after’ : Absorbance of undiluted bacterial culture after co-culture with plants
00_raw_data/dat/pixel_data/[.]/ : Each directory represents one plate of plants, which was processed with our script ‘process_athal_scans.py’. Directories are named using the formula: 'processed_[experiment]__[plate]', where the [experiment] refers to the unique experiment identifier of each batch, and [plate] refers to the unique plate identifier of each scanned image. These values match 'experiment' and 'plate' variables in the metadata files. Within each directory, there are two files:
- image_masked.png : image of plants after application of a background filter (“mask”)
- pixel_data.txt : extracted data from images. NA = Not available because the image processing software could not extract or calculate pixel data from image. Data column descriptions are as follows:
- 'subimage' : Name of individual plant images (default is image divided into 12x8 subimages)
- 'all_plant_pixels' : total number of pixels identified as plant (using a ratio of green to red and blue colour intensity; see methods), as opposed to background. This value will change depending on the resolution of the image.
- 'ratio_[r/g/b]_q5' : lower 5th percentile value of ratio of red (r), green (g) or blue (b) pixel intensity to total pixel intensity (e.g. r/(r+g+b) ) across the entire plant area
- 'ratio_[r/g/b]_q95' : upper 95th percentile value of ratio of red (r), green (g) or blue (b) pixel intensity to total pixel intensity (e.g. r/(r+g+b) ) across the entire plant area
- 'ratio_[r/g/b]_median' : median (50th percentile) value of ratio of red (r), green (g) or blue (b) pixel intensity to total pixel intensity. (e.g. r/(r+g+b) ) across the entire plant area
- 'pix_[r/g/b]_median' : median (50th percentile) raw pixel colour intensity across entire plant area for red (r), green (g) or blue (b)
- 'diff_[rb/gb/gr]_median]' : median (50th percentile) of differences between median colour values of two colour channels. 'rb' = difference between red and blue intensity; 'gb' = difference between green and red intensity; 'gr' = difference between green and red intensity.
00_raw_data/meta/ : Directory containing all metadata files associated with data outputs. Metadata filenames correspond to the data ‘Experiment’ column.
- [.]_meta.csv : Column descriptions are as follows:
- 'experiment' : experiment name, unique for each temporal experimental trial.
- 'plate' : plate name, unique across all plates in the experiment.
- 'row' and 'col' : row and column labels for well.
- 'well' : combination of 'row' and 'col'.
- 'protect' : candidate protective strain, either MOCK, WCS365, CHA0, CH267, or Pf5. In some experiments, we used extra wells to run pilot tests for other strains but these strains were not included in our final manuscript.
- 'path' : pathogenic strain, either MOCK or N2C3. A label of 'L' meant the isolate had a LacZ genomic insert, and a label of 'neon' or 'crim' designated which (if any) fluorescent protein marker that strain included. In some experiments, we used extra wells to run pilot tests for other strains but these strains were not included in our final manuscript.
- 'plant' : plant genotype, either col0, bbc, bik1, or nej. nej plants were not used in analysis.
- 'protect_od' : Optical density (600nm, 100uL) of candidate protective strain at time of inoculation.
- 'path_od' : Optical density (600nm, 100uL) of candidate pathogenic strain at time of inoculation.
- 'date_ster' : Date (YYYY-MM-DD) that seeds were sterilized and plants began cold-stratification process.
- 'date_germ' : Date (YYYY-MM-DD) that plants were germinated on sterile plant media.
- 'date_inoc': Date (YYYY-MM-DD) that bacteria were inoculated on plants in 96-well hydroponic system.
- 'date_process' : Date (YYYY-MM-DD) that plants were processed at end of experiment.
- 'light_dn' : Day / night hours for light.
- 'light_lx' : Light intensity plants were exposed to for experiment, units lux.
- 'temp_dn' : Temperature (C) during day/night.
- 'MS' : Murashige and Skoog concentration used in plant media, where 1 is full strength.
- 'MES' : MES buffer concentration used in plant media, where 1 is full strength.
- 'pH' : pH of plant media.
- 'note' : experiment-specific notes about that particular plant or treatment
02_Analysispath_fluoranalysis
This includes data and metadata from MYCroplanter experiments used to determine the effect of inoculation ratio on both bacterial competition outcome and plant health outcome. This data were used in Figures 4 and 7a, and Supplementary Figure S10.
00_raw_data/dat/fluor_data/ : Fluorescence readings from plates after co-inoculation of bacteria and plants in MYCroplanters. Plants were removed prior to reading. CFU counts were also done for a subset of wells to correlate fluorescence readings back to “real” CFU counts.
- Data-Trial-[.]-MYC.csv : Fluorescence data from experimental plants inoculated with different bacterial inoculants. Column descriptions are as follows:
- ‘plate_alias’ : Plate name. Fluorescence data was collected by a different researcher who generated scanned plant images and metadata files, so 'plate alias' refers to the unique plate identifiers given to all fluorescence data, whereas 'plate' refers to the unique plate identifier of scanned plant images. Plate alias is therefore used to cross-reference fluorescence data plate data with other plate data.
- ‘row’ and ‘col’ : Row and column, respectively, of 96-well plate
- ‘Neon-raw’ : Raw fluorescence reading of well for the “Neon” fluorophore (read from top)
- ‘Crim-raw’ : Raw fluorescence reading of well for the “Crimson” fluorophore (read from top)
- ‘Neon-cfu’ : CFU count (per mL) of strain containing the ‘Neon’ plasmid (Not done for all wells)
- ‘Crim-cfu’ : CFU count (per mL) of strain containing the ‘Crimson’ plasmid (Not done for all wells)
- ‘Neon-100’ and ‘Crim-100’: Raw fluorescence reading of a 100uL aliquot of a well. This was done once in our third data trial to determine whether there was substantial effect of well volume on fluorescence readings. Once plants are removed, there tends to be some variation in well volume depending on the size and ‘thirsty-ness’ of the plant. Thus, we compared full well and 100uL subsets of wells to see how much fluorescence readings varied between wells of slightly different volumes.
- ‘Neon-btm-raw’ and ‘Crim-btm-raw’ : Raw fluorescence reading of whole well for ‘Neon’ and “Crimson’ fluorophores, respectively, using the ‘read-from-bottom-of-well’ setting in our plate reader. This was done once to confirm that readings from the top and bottom of wells were well correlated. We found that top readings had a slightly better range of fluorescence readings, so top measurements only were used for the remainder of our experiments.
- fluor_data.csv : A set of plants inoculated with bacteria that were measured from top and bottom, and also had their absorbance measured. This was done as validation to confirm that top and bottom fluorescence readings were roughly equivalent. Column descriptions are as follows:
- ‘plate_alias’ : Plate name. Fluorescence data was collected by a different researcher who generated scanned plant images and metadata files, so 'plate alias' refers to the unique plate identifiers given to all fluorescence data, whereas 'plate' refers to the unique plate identifier of scanned plant images. Plate alias is therefore used to cross-reference fluorescence data plate data with other plate data.
- ‘row’ and ‘col’ : Row and column, respectively, of 96-well plate
- 'plate_well' : Plate alias, row and column concatenated
- 'Abs600' : Absorbance of whole well at 600nm at end of experiment. Used to compare to fluorescence to validate that turbid wells also had higher fluorescence.
- ‘Neon-raw’ : Raw fluorescence reading of well for the “Neon” fluorophore (read from top)
- ‘Crim-raw’ : Raw fluorescence reading of well for the “Crimson” fluorophore (read from top)
- ‘Neon-btm-raw : Raw fluorescence reading of well for the “Neon” fluorophore (read from bottom)
- ‘Crim-brm-raw’ : Raw fluorescence reading of well for the “Crimson” fluorophore (read from bottom)
00_raw_data/dat/pixel_data/
- Same as in 01_Analysispath_compratio/dat/pixel_data; please refer to descriptions of outputs above
00_raw_data/meta/ : directory containing all metadata files associated with data outputs. Metadata filenames correspond to the data ‘Experiment’ column.
- [.]_meta.csv : Column descriptions are as follows:
- 'experiment' : experiment name, unique for each temporal experimental trial.
- 'plate' : plate name, unique across all plates in the experiment. This will match the plate names in pixel_data
- 'row' and 'col' : row and column labels for well.
- 'protect' : candidate protective strain, either MOCK, WCS365, CHA0, CH267, or Pf5. A label of 'L' meant the isolate had a LacZ genomic insert, and a label of 'neon' or 'crim' designated which (if any) fluorescent protein marker that strain included.
- 'path' : pathogenic strain, either MOCK or N2C3. A label of 'L' meant the isolate had a LacZ genomic insert, and a label of 'neon' or 'crim' designated which (if any) fluorescent protein marker that strain included.
- 'plant' : plant genotype, either col0, bbc, bik1, or cerk1.
- 'protect_od' : Optical density (600nm, 100uL) of candidate protective strain at time of inoculation.
- 'path_od' : Optical density (600nm, 100uL) of candidate pathogenic strain at time of inoculation.
- 'date_ster' : Date (YYYY-MM-DD) that seeds were sterilized and plants began cold-stratification process.
- 'date_germ' : Date (YYYY-MM-DD) that plants were germinated on sterile plant media.
- 'date_inoc': Date (YYYY-MM-DD) that bacteria were inoculated on plants in 96-well hydroponic system.
- 'date_process' : Date (YYYY-MM-DD) that plants were processed at end of experiment.
- 'light_dn' : Day / night hours for light.
- 'light_lx' : Light intensity plants were exposed to for experiment, units lux.
- 'temp_dn' : Temperature (C) during day/night.
- 'MS' : Murashige and Skoog concentration used in plant media, where 1 is full strength.
- 'MES' : MES buffer concentration used in plant media, where 1 is full strength.
- 'pH' : pH of plant media.
- 'plate_alias' : alternative plate name that was used to cross-reference to plant images to fluorescence data. An alternative plate name was used because fluorescence data was collected by a different co-author, who used her own set of naming schemes for fluorescence data.
- 'note' : experiment-specific notes about that particular plant or treatment
03_Analysispath_priorityeffects
This includes data and metadata from MYCroplanter experiments used to determine the effect of inoculation ratio and priority effects on both bacterial competition outcome and plant health outcome. This data were used in Figures 5, 6, and 7b, and Supplementary Figure S11.
00_raw_data/dat/cfu_data/ : CFU counts for strain 1 in priority effects experiments found in Figure 6. Plants were inoculated with first strain, and then random plants were destructively sampled to determine the approximate CFU load on roots after a certain about of time had passed. Roots were beat with beads to dislodge bacteria and resulting mixture was plated in serial dilution to determine CFU counts.
- [.].csv : These file names do not follow any naming convention. Column descriptions are as follows:
- ‘experiment’ : Experiment name, which matches names in metadata
- 'transfer_h' : hours after inoculation that plants were transferred to strain 2 (or in this case, time of destructive sampling to count CFUs from strain 1)
- ‘strain’ : which strain; WCS365 or N2C3
- ‘fluor’ : fluorescent marker found in the strain being counted
- 'plant' : plant genotype, all col0
- 'germ_plate' : If available, which batch of germination plates plant came from.
- 'strain_number' : strain 1 or 2.
- 'abs_600' : absorbance at 600nm of second strain at inoculation
- 'bio_rep' : biological replicate (refers to different plants / wells)
- 'tech_rep' : technical replicate (refers to different aliquots from ground root suspension)
- 'CFU' : CFU count per plate (raw)
- 'dilution' : dilution factor, 10^x
- 'mL_plated' : how many mL of dilution were plated on dilution plate
- 'total_volume' : total volume of original cell suspension (uL)
- 'CFU_mL' : CFU per mL calculation based on previous column data.
- 'CFU_well' : CFU per well (or per plant)
00_raw_data/dat/fluor_data/ : Fluorescence readings from plates after co-inoculation of bacteria and plants in MYCroplanters in priority effects experiments. Plants were removed prior to reading. CFU counts were also done for a subset of wells to correlate fluorescence readings back to “real” CFU counts.
- long_data_p_effect_[.]_MYC.csv : NA and an empty cell both mean no data collected. Column descriptions are as follows:
- ‘plate’ : Plate name. Matches metadata column of same name.
- ‘experiment’ : Experiment name. Matches metadata column of same name.
- ‘row’, ‘col’, and ‘well’ : Row, column, and well, respectively, of 96-well plate
- ‘abs600nm’ : In some experiments, the total absorbance of the well at 600nm was read to confirm that cell density corresponded roughly to increases in fluorescence.
- ‘neon_top’ and ‘neon_btm’ : Raw fluorescence reading of well for the “Neon” fluorophore (read from top and bottom, respectively). After preliminary analysis, we found that top readings had a slightly larger range of fluorescence values so only top readings were used for analysis.
- ‘crim_top’ and ‘crim_btm’ : Raw fluorescence reading of well for the “Crimson” fluorophore (read from top and bottom, respectively). After preliminary analysis, we found that top readings had a slightly larger range of fluorescence values so only top readings were used for analysis.
- ‘readtime’ : Time (in days) after inoculation that fluorescence was read from plates. In one experiment, fluorescence was also read at time 0 to validate that dilutions had been prepared properly.
00_raw_data/dat/pixel_data/
- Same as in 01_Analysispath_compratio/dat/pixel_data; please refer to descriptions of outputs above
00_raw_data/meta/ : directory containing all metadata files associated with data outputs. Metadata filenames correspond to the data ‘Experiment’ column.
- [.]_meta.csv : Column descriptions are as follows (not all metadata files have all columns):
- ‘experiment' : experiment name, unique for each temporal experimental trial.
- 'plate' : plate name, unique across all plates in the experiment.
- 'row' and 'col' : row and column labels for well.
- 'protect' : candidate protective strain, either MOCK, WCS365, CHA0, CH267, or Pf5. A label of 'L' meant the isolate had a LacZ genomic insert, and a label of 'neon' or 'crim' designated which (if any) fluorescent protein marker that strain included.
- 'path' : pathogenic strain, either MOCK or N2C3. A label of 'L' meant the isolate had a LacZ genomic insert, and a label of 'neon' or 'crim' designated which (if any) fluorescent protein marker that strain included. Note that datasets sometimes include extra strains that were part of the experiment but not used in this manuscript or analysis.
- 'transfer_h' : hours after inoculation that plants were transferred to strain 2. An “hourly” transfer time of 0.01 was chosen to represent the ‘dip’ treatment.
- 'strain0', 'strain1' and 'strain2': in cases where strains were introduced at different time points, strain1 and and strain2 indicate which strain was introduced first and second, respectively. 'strain 0' column was only used to specify cases where plants were dipped in sterile solution first ('MOCK') before being dipped in strain 1. This was used as a control, as we suspected that wet roots might 'absorb' more inoculation media than wet roots. Column values could be either N2C3, WCS365, NA (removed from experiment), MOCK (no bacteria media control), or None (no dipping in second liquid at all). A label of 'L' meant the isolate had a LacZ genomic insert, and a label of 'neon' or 'crim' designated which (if any) fluorescent protein marker that strain included.
- 'plate_filled_with' : In priority effects experiment, which strain the plate was filled with. In our first priority effects experiment, plants were dipped in strain 1 but incubated in a plate of strain 2, so this column will be identical to strain 2. However, for our second prioirty effects experiment we dipped plants in strain 1, then dipped plants in strain 2, and incubated plants in a fresh plate without bacteria in it; therefore, this column would contain ‘MOCK’.
- 'plant' : plant genotype, either col0, NA, or NOPLANT. The designation 'NOPLANT' is used in priority effects experiments to indicate that plants were transferred out of this plate after a set amount of time, so this well should not be used in analysis. However, we did use this well to check for potential contamination issues.
- 'strain1_od': Optical density (600nm, 100uL) of strain 1 at time of inoculation.
- 'strain2_od': Optical density (600nm, 100uL) of strain 2 at time of inoculation.
- 'date_ster' : Date (YYYY-MM-DD) that seeds were sterilized and plants began cold-stratification process.
- 'date_germ' : Date (YYYY-MM-DD) that plants were germinated on sterile plant media.
- 'date_inoc': Date (YYYY-MM-DD) that bacteria were inoculated on plants in 96-well hydroponic system.
- 'date_process' : Date (YYYY-MM-DD) that plants were processed at end of experiment.
- 'light_dn' : Day / night hours for light.
- 'light_lx' : Light intensity plants were exposed to for experiment, units lux.
- 'temp_dn' : Temperature (C) during day/night.
- 'MS' : Murashige and Skoog concentration used in plant media, where 1 is full strength.
- 'MES' : MES buffer concentration used in plant media, where 1 is full strength.
- 'pH' : pH of plant media.
- 'Type' : In priority effects experiment, used to differentiate treatments where plants were transferred from strain 1 to strain 2 ('Priority'); dipped in both strains then transferred to a fresh plate (‘doubledip’); plants that were used for simultaneous inoculation controls ('MIX'); or single-strain inoculations ('Single').
- ‘plant_quality’ : An internal note of whether plant roots were long enough to be seen at time of transfer. Plant roots were sometimes so short they were barely visible at time of transfer from strain 1 to strain 2, so we were worried no bacteria would be able to colonize the root before plants were transferred to the second strain. We ended up using all plants in our analysis, so this column became obsolete.
- ‘Incubator_germ’ : An internal note of whether plants were germinated in a different chamber than the rest of experiments. During our experiments, one of our co-authors changed lab spaces and we conducted some plant experiments in a different room.
- ‘Incubator_growth’ : An internal note of whether plants were incubated with bacteria in a different chamber than the rest of experiments. During our experiments, one of our co-authors changed lab spaces and we conducted some plant experiments in a different room.
04_Analysispath_priorityplates
Raw data associated with all solid phytoagar plant cultivation experiments. This includes plant weight information and CFU counts from root-associated bacteria; as well as CFU counts from in vitro competition experiments. Used for Supplemental Figure S13
00_raw_data/ : Data from solid phytoagar plate experiments
process_athal_2022-04-19.csv : Plant weight data transcribed from solid phytoagar plant cultivation experiments. Column descriptions are as follows:
- 'date_processed' : Day the data was collected (end of experiment)
- 'plate' : plate ID; not unique across experiments
- 'treatment' : Bacterial inoculation treatment. W = WCS365, WL = WCS365 with lacZ genomic insertion; N = N2C3, NL = N2C3 with lacZ genomic insertion; 0 = MOCK. NS, NP, and NQ are N2C3 strains with syr:syp, ppi1:ppi2, and syr:syp:ppi1:ppi2 deletions, respectively, which were not used in analysis. Strains are separated with '+' (simultaneous inoculation) or '++' (delayed inoculation of second strain).
- 'weight_mg' : weight of plant in milligrams.
- 'strain1' : First strain inoculated (if delayed inoculation).
- 'strain2' : Second strain inoculated (if delayed inoculation).
- 'delay' : description of what kind of inoculation delay was used. '-' means no delay, '++' means 24 hour delay.
2022-08-07_platecount_invivo_comp_lacz.csv : CFU count data from plants that were weighed in process_athal_2022-04-19.csv. Column descriptions are as follows:
- 'plate' : plate ID; cross-reference to 'plate' column in process_athal_2022-04-19.csv.
- 'treatment' : Bacterial inoculation treatment. W = WCS365; WL = WCS365 with lacZ insertion; N = N2C3; NL = N2C3 with lacZ insertion. NS, NP, and NQ are N2C3 strains with syr:syp, ppi1:ppi2, and syr:syp:ppi1:ppi2 deletions, respectively, which were not used in analysis.
- 'strain1' : First strain inoculated (when delayed inoculation)
- 'strain2' : Second strain inoculated (when delayed inoculation)
- 'delay' : '-' indicates simultaneous inoculation and '++' indicates 24h delay in second strain inoculation.
- 'blue' : number of blue colonies
- 'white' : number of white colonies
- 'Number_of_plants_pooled' : number of plants pooled in bead-beating tube before serial dilution to get CFU counts
- 'dilution' : number of 10-fold dilutions done to get CFU counts in 'blue' and 'white' (e.g. 2 = 10^-2)
2022-08-07_platecount_noplantcom_lacz.csv : CFU count data from in vitro competition between strains with and without lacZ insertion. Bacterial strains were dotted onto plant media, either simultaneously (mixed and dotted) or in sequence (second strain dotted after 24h delay).
- 'date_process' : Date of data collection (end of experiment)
- 'plate' : Plate ID, different from 'plate' in other two files
- 'row' and 'col' : position in the plate that colonies were dotted
- 'blue' : Number of blue colonies
- 'white' : Number of white colonies
- 'dilution': Dotted bacteria were washed with a small amount of liquid and then serially diluted to get CFU plate counts. Dilution factor is 10-fold (e.g. 2 = 10^-2)
00_meta/ : Metadata files for in vitro and in vivo priority effect experiments on square agar plates.
- inoc_dot_plates.csv : Column descriptions are as follows:
- ‘plate’ : Plate identifier where cultures were dotted
- ‘row’ and ‘col’ : row and column in square plate; dimensions are A-F x 1-6
- ‘strain1’ : First strain used (W = WCS365, WL = WCS365 with LacZ insertion, N = N2C3, NL = N2C3 with LacZ insertion)
- ‘strain2’ : Second strain used (W = WCS365, WL = WCS365 with LacZ insertion, N = N2C3, NL = N2C3 with LacZ insertion)
- ‘ratio1’ : Relative amount of strain 1
- ‘ratio2’ : Relative amount of strain 2
- ‘od1’ : Optical density (Absorbance at 600nm, 1cm cuvette)
- ‘od2’ : Optical density (Absorbance at 600nm, 1cm cuvette)
- ‘volume1’ : Volume (uL) of strain1 culture used (after dilution to od1)
- ‘volume2’ : Volume (uL) of strain2 culture used (after dilution to od2)
- ‘date_inoc’ : Date of inoculation
- ‘date_process’ : Date of sampling (end of experiment)
- inoc_plant_plates.csv : Column descriptions are as follows:
- ‘plate’ : Plate identifier where plant was grown
- ‘position’ : On plates, plants are grown either on the left, middle, or right. Positions ‘a’, ‘b’, and ‘c’ refer to these locations, respectively.
- ‘strain1’ : First strain inoculated (W = WCS365, WL = WCS365 with LacZ insertion, N = N2C3, NL = N2C3 with LacZ insertion, ‘mock’ = no bacteria control)
- ‘strain2’ : Second strain inoculated (W = WCS365, WL = WCS365 with LacZ insertion, N = N2C3, NL = N2C3 with LacZ insertion,, ‘mock’ = no bacterial control)
- ‘delay’ : Whether or not there was a delay between strain1 and strain2 inoculation. ‘-’ means no delay, ‘++’ means 24 hour delay.
- ‘od1’ : Optical density (Absorbance at 600nm in 1cm cuvette) of first strain
- ‘od2’ : Optical density (Absorbance at 600nm in 1cm cuvette) of second strain
- ‘volume1’ : Volume (uL) of strain 1 culture used (after dilution to od1)
- ‘volume2’ : Volume (uL) of strain 2 culture used (after dilution to od2)
- ‘test’ : Experiment type’ should be all ‘delay’ because these data were used to test delay effects
- ‘date_inoc’ : Date plants were inoculated
- ‘date_transplanted’ : Date plants were transplanted from original germination plates (with sucrose) to experimental plates (no sucrose)
- ‘date_sow’ : Date plants were germinated on germination plates
- date_processed’ : Date plants were sampled
05_SupplementalData
Data used for supplemental figures S14 and S15.
Plasmid_burden/ : Data from growth curve experiments to test for plasmid burden (Supplemental Figure S14)
- 001_Growth_curve_data_compiled_24-03-29.csv : NA = Not applicable; well not used in experiment. Absorbance and fluorescence of cultures over time to assess for growth defects in different strains. Column descriptions are as follows:
- 'file' : Unique identifier for original file where data is stored
- 'Time' : Time (in hours) of measurements
- ‘Well’ : Well in 96-well plate (A1-H12)
- ‘Abs600’ : Absorbance at 600nm of culture
- ‘Name’ : Full name identifier of strain in well. Can be either N2C3 (WT), WCS365 (WT), or WCS365-LacZ (with genomic lacZ insertion) with plasmids Neon or Crim. NA and Blank both refer to media blanks.
- ‘Strain’ : Strain of bacteria, can be either N2C3 (WT), WCS365 (WT), or WCS365-LacZ (with genomic lacZ insertion)
- ‘Plasmid’ : Which fluorescent plasmid the strain contains, either Neon, Crim, or NA
- ‘Media’ : Media type, either LB (Luria broth) or 0.5MS0.5MES30mMSuccinate (plant growth media with succinate for a carbon source)
- ‘Experiment’ : Which temporal experiment / batch
- ‘LacZ’ : Whether the strain contained a genomic LacZ insertion
- ‘Innoculation’ : If NA, then inoculated at 0.01 Abs600. Inoculation density otherwise noted.
Plasmid_loss/ : Data from experiment to determine rate of plasmid loss between different strain and plasmid combinations. Used for Supplemental Figure S15.
- Plasmid-Maint-2-Nov-2023.csv : Strains WCS365 and N2C3 with and without crimson and neon plasmids were grown with plant roots for 7 days, and then plated onto non-selective media (LB). CFUs were selected from the plate with non-selective media and dotted onto a plate with selective antibiotics (LB + tetracyclin) to determine which colonies still contained the plasmid (which conferred tet resistance). Column descriptions are as follows:
- 'strain' : Which strain was tested; either WCS365 (WT, no plasmid), N2C3 (WT, no plasmid), WCS365-Crim, WCS365-Neon, N2C3-Crim, or N2C3-Neon.
- 'dotted ' : Number of colonies chosen from non-selective media plates that were dotted onto selective media plates after 7 days of strain growth on plant roots.
- 'survived' : Number of colonies that survived after being dotted onto selective media plates (LB with tetracyclin).
- 'Rep' : Technical replicate. NA refers to "Not applicable" because CFUs were plated from the same overnight culture on day 0 of the experiment.
- 'Experiment' : Two trials were done of this experiment, 1 and 2.
- 'Day' : As a control, the initial starting cultures were also plated, dotted, and counted to determine the starting frequency of cells with the plasmid. The experiment ended on day 7.
HEALTHSCORE_DATA
Human assessments of plant health compared to algorithm assessments. This data was used to determine whether the Health Scores calculated from scanned plant images were reasonable approximations of plant health, according to human experts.
- assessments/ : Individuals ranked plant health of 10 sets of 10 plants. Each PDF is one person's health score assessment. Individuals were asked to move images of plants in order of "most sick" (left) to "least sick" (right), and also delineate (with vertical line) where they felt the cutoff for "healthy" and "unhealthy" was. Files were converted from pptx to PDF.
- individual_plants/ : Plant images chosen at random (random number generator) for plant health score assessments. A total of 84 plants were used because some images were repeated across multiple "sets" of plants as a validation measure to see whether the same person perceived plants consistently as "health" or "not healthy" across different sets
assessment_data_anonymized.csv : Manually entered data from assessments/, where individual humans are anonymized (HUMAN1, HUMAN2). Empty cells are when data was not available, as we found afterwards there was a missing plant image in the first set of ten. Metadata columns are as follows:
- 'anonymised_person' : Initials of researcher who made the assessment
- 'set' : which set of plants (corresponds to ppt slide)
- 'rank' : rank (1-10) of eacg plant given by researcher
- 'id' : plant ID, which is found in the top left corner of each plant image
- 'healthy' : whether each plant was considered 'healthy' (right of vertical line) or 'not healthy' (left of vertical line) by researcher. T = True, which means plants were considered healthy. F = False, which means were not considered healthy.
healthscore_validation.pdf : File given to researchers for assessments. Researchers were given a PPTX (or ODP) file, but we have converted it to a PDF here. Researchers were asked to click and drag plant images so that they were ordered least healthy (left) to most healthy (right); and to move the vertical bar between where they considered plants 'healthy' (right) and 'not healthy' (left). Researchers then saved their rankings and returned their files, which were saved in assessments/.
MYCROPLANTER_3DPRINTING
We have included .stl files for all 3D printer components of our system, including a PDF diagram showing what each piece should look like and how they fit together in our "MYCROPLANTER_3DPRINTING" file collection. If you are printing MYCroplanters yourself, we advise that you test print a single MYCroplanter (mycroplanter_1.stl) before committing to a set of 96. Note that depending on the quality of your print, you may need to pierce the bottom of the MYCroplanter funnel with a small needle to ensure there is a hole. The hole diameter is meant to be 0.4mm, and some printers cannot print accurately at that resolution.
- DIAGRAME_mycroplanters_all_components.pdf : Diagram for all MYCroplanter system components
- germination_tray.stl : 3D printing file for germination tray, which allows A. thaliana seeds to germinate on semi-solid phytoagar with plant media. MYCroplanters are placed in each divot; seeds are dotted at the bottom of each MYCRoplanter.
- lid_brace.stl : 3D printing file for brace to hold lid of 96-well plate up. Allows more space for plants to grow vertically.
- mycroplanter_1.stl : 3D printing file for single MYCroplanter (for test prints)
- mycroplanter_96.stl : 3D printing file for a set of 96 MYCroplanters
- scanning_tray_base.stl : 3D printing file for the bottom half of the scanning tray, which is used to scan plants after experiments
- scanning_tray_top.stl : 3D printing file for top half of scanning tray, which is used to can plants after experiments
Sharing/Access information
Processed versions of the data and code associated with this project can be found on our GitHub
Software
MYCROPLANTER_PROCESSING_TUTORIAL
- This collection of scripts and instructions contains code required to run the processing script for scanned plant images from the MYCroplanter system. Please note that this includes the most recent version of the script, 'process_athal_scans.py' as of 2024-04-24. For the most up-to-date versions, always refer to our Github.
- Data were processed using code found on our GitHub:
Methods
This dataset includes all raw data generated from MYCroplanter experiments between January 2021 through March 2024. We have also included processing code (for image processing) and STL files for 3D printing MYCroplanters.
MYCroplanter data (01_Analysispath_compratio, 02_Analysispath_fluoranalysis, 03_Analysispath_priorityeffects)
MYCroplanters are custom 3D printed polylactic acid (PLA) micro-planters that grow Arabidopsis thaliana seedlings in 96-well plates. Sterile MYCroplanters (sterilized with chlorine gas) were placed in a 8x12 array germination tray on top of solid plant media containing 1/2X MS, 0.5% MES, 2% sucrose, 0.5% phytoagar, pH 5.8 (pH adjusted with 1M KOH). The germination tray were placed inside a standard single-well polystyrene plate. Sterile A. thaliana seeds (sterilized with 0.3% H2O2 + 70% EtOH and cold-stratified in advance for 2-5 days) were individually pipetted into each MYCroplanter such that seeds rest directly on the phytoagar below the mycroplanter. Plants were allowed to germinate between 5-7 days.
At 5-7 days old, seedlings were inoculated with bacteria by transferring each mycroplanter into a 96-plate well pre-filled with 1/2X MS plant growth media and bacterial inoculant. Plates were sealed with 3M micropore tape and plants are kept in bacterial solution for 7 days. Growth conditions were approximately 110 µMol light in 12h/12h day/night conditions, at 23C.
To collect plant health data, MYCroplanters were transferred to a custom 3D printed scanning tray at the end of the experiment, covered in a piece of glass, and inverted onto a Epson Perfection V850 Pro Photo Scanner. Reflective scans in 48-bit colour at a resolution of 600dpi were taken. Images were processed using a custom python script, which divided each scanned image into an 8x12 array and separated plants from the background using a green and red ratio filter. Fluorescence readings (where applicable) for mCrimson and Neon labelled bacterial strains were measured by a plate reader at 612/646 and 495/525 nm (excitation/emission), respectively. Cell counts (where applicable) were derived from serially diluting cultures from wells or plants (homogenized in MgSO4 buffer) and growing on LB or LB + antibiotics with X-gal. To differentiate strains, some strains had a lacZ genomic insertion.
Plate data (04_Analysispath_priorityplates)
A. thaliana seedlings were sterilized using chlorine gas (300mL 6% bleach + 6mL HCl) for one hour and cold-stratified in water for 2 days. Seeds were germinated on vertical 1% phytoagar plates with 1/2MS 1/2MES 2% sucrose for 6 days. Seedlings were transferred to 1/2MS 1/2MES (no sucrose) 1% phytoagar plates, and allowed to acclimatize for one day before inoculations. For concurrent inoculations, 6µL total of 0.001OD cultures were added to the plant. For delayed inoculations, 6µL of 0.001 OD culture was inoculated on the first day, and 6µL of 0.001OD culture was inoculated on the second day. For in vitro treatments, cultures were spotted onto identical phytoagar plates with no plants growing. In each treatment, one of the strains included a lacZ genomic insertion. Relative abundances of CFUs were counted from serially diluted plates.
Supplemental data (05_SupplementalData)
A full description of methods can be found in our manuscript. In short, strains were tested for plasmid burden (via growth curves) and for plasmid loss (via CFU counts on selective and non-selective media). For growth curves, strains were grown in either LB (Luria broth) or plant growth media + succinate (0.5MS 0.5MES +succinate) at 28C and absorbance at 600nm was measured every 15 minutes to estimate increases in cell density. For plasmid loss experiments, strains were inoculated onto 5-day-old seedlings in MYCroplanters and incubated for 7 days. After 7 days, liquid cultures were spread onto non-selective media plates (LB) and individual CFUs were transferred to selective media plates (LB + tetracyclin) to determine what percentage of CFUs retained the plasmid (which conferres tetracyclin resistance).
Health score data (Healthscore_data)
Individuals ranked plant health of 10 sets of 10 plants by clicking and dragging plant images so that they were ordered least healthy (left) to most healthy (right); and by moving the vertical bar between where they considered plants 'healthy' (right) and 'not healthy' (left). Researchers then saved their rankings and returned their files. This data was used to determine whether the Health Scores calculated from scanned plant images were reasonable approximations of plant health, according to human experts.