loading

Experiments

Here we present to you an overview of the experiments our team conducted in the experimental field and in the laboratory.

Figure 1.Overview of the team's experiments.

Transplanting tomato plants & Soil impregnation with PGPB

The team decided on the bacterial mixes for the impregnation of the soil. These contained plant growth-promoting bacteria (PGPB) and zeolite and biochar as carriers ( pH 6.8-7.2).The carriers used resulted in solid and liquid mixes (Table 1).

Table 1.Plot number and corresponding mix used for soil impregnation (carrier and strain).
Plot Carrier Strain
B1 Zeolite B. amyloliquefaciens, subgroup B. subtilis strain RS-3
B2 Liquid B. thuringiensis, subgroup B. cereus strain 109/18
B3 Liquid B. subtilis, subgroup B. subtilis strain 548
B4 Biochar Control
B5 Zeolite Control
B6 Liquid B. subtilis, subgroup B. subtilis strain Z3
B7 Biochar B. mojavensis, subgroup B. subtilis strain 5B2
B8 Zeolite B. mojavensis, subgroup B. subtilis strain 5B2
B9 Zeolite B. subtilis, subgroup B. subtilis strain 557
B10 Biochar B. amyloliquefaciens, subgroup B. subtilis strain RS-3
B11 Liquid B. amyloliquefaciens, subgroup B. subtilis strain RS-3
B12 Liquid B. subtilis, subgroup B. subtilis strain 557

Therefore, transplanting and soil impregnation with the selected microorganisms took place in the experimental field. More specifically, young tomato plants, approximately 10 days old, from certified seeds of Solanum lycopersicum L. cv. Rio Grande (HEINZ 1015 hybrid), were transplanted and the space between plants in the row was 50 cm, as the experimental field was sectioned into 12 plots. Our experiment followed a completely randomized design, with 12 different treatments (10 with PGPB and 2 control), as seen in the previous table. The mixes of the PGPB which we used, were diluted with tap water, in order to achieve the desired concentration of bacteria, which was 7*107 CFU/ml. These were added to the soil close to the tomato plants and the application rate of PGPB at the corresponding plot was 7 lt/ha.

Soil sample collection

Soil samples were collected manually, in depth of 0-30 cm from the surface at the experimental field, with Greiner 15 ml falcon tubes, in order to proceed later on with the metagenomic analysis. During sample collection, all the members that participated wore gloves, which they changed after the collection of each sample, to avoid cross-contamination. These samples were properly stored in the freezer in liquid nitrogen to ensure their quality, till the beginning of our experiments.

DNA extraction & 16S rRNA sequencing

The bacterial diversity in 12 different soil samples from the plots B1-B12 was analyzed, by exploring the bacterial 16S rRNA gene diversity using molecular techniques (Illumina MiSeq).

DNA extraction was performed on the soil samples collected previously, using the Macherey-Nagel™ NucleoSpin™ kit. Afterwards, 16s rRNA gene sequencing was performed, using Illumina technology with a MiSeq PE300 sequencer, following the manufacturer's recommendations.

Primers S-DBact-0341-b-S-17 (CCTACGGGNGGCWGCAG) and S-D-Bact-0785-a-A-24 (GACTACHVGGGTATCTAATCC)1 were chosen for 16S rRNA genes sequencing. The 16S rRNA gene of Escherichia coli K12 was used as the positive control and template for the construction of standard curves.

Also, quantitative PCR (qPCR) was performed in a StepOne system (Applied Biosystems; Thermo Fisher Scientific). Triplicate samples were used to quantify rRNA gene copy number for the soil samples using 16S specific primers. BACT1369F CGGTGAATACGTTCYCGG and PROK1492R GGWTACCTTGTTACGACTT2. A three-step amplification procedure, including denaturation, annealing and extension, was performed. Analysis of the data collected from sequencing followed, using mothur 1.44.1 version and SILVA 138 database for alignment and taxonomic classification.

Here, you can find the exact protocols that we followed.

Data analysis and Machine Learning model

Having prior knowledge regarding the format of our data, we worked on the implementation of our final product, an algorithm, by writing scripts in programming language R (version 4.2.0). R helped us in the rapid development of code for different classifiers.

Analysis of the data collected from sequencing took place, using mothur 1.44.1 version and SILVA 138 database for alignment and taxonomic classification. After formatting our data to a final tsv file, we found the best performing model in terms of speed and accuracy and run hyperparameter tuning to find the optimal model-specific parameters.

References
  1. Klindworth, A., Pruesse, E., Schweer, T., Peplies, J., Quast, C., Horn, M., & Glöckner, F. O. (2013). Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic acids research, 41(1), e1. doi: 10.1093/nar/gks808
    Full text
  2. Suzuki, M. T., Taylor, L. T., & DeLong, E. F. (2000). Quantitative analysis of small-subunit rRNA genes in mixed microbial populations via 5'-nuclease assays. Applied and environmental microbiology, 66(11), 4605-4614. doi:10.1128/AEM.66.11.4605-4614.2000
    Full text