GENEcision Example · illustrative data
← Back to site
Interactive decision report

Strain-pool selection for a target crop

Client: AgTech bio-input developer (illustrative) Pool: 12 candidate isolates + 3 commercial benchmarks Report: GC-R-0142 · v1.0

Executive summary

Decision scenario

Change the scenario and every section updates.
Target crop
Goal

Candidate ranking

Estimated probability that each strain improves the chosen goal in greenhouse and field trials. Select a row to see its profile.

#StrainProbabilitySafetyDecision

Best consortia

Combinations of 2–3 compatible strains that cover each other's gaps.

Functional trait matrix

lowhigh

Genome-derived trait scores (normalized gene-cluster frequencies), ordered by the current ranking. Benchmarks appear at the bottom.

Biocontrol potential

What the client receives

  • 01_Executive_summary.pdf2 pages, plain language
  • 02_Decision_report.htmlthis interactive file
  • 03_Technical_report.pdfmethods, QC, figures, limitations
  • genomes/*.fasta · *.gbk · *.gff3assemblies and annotations
  • tables/trait_matrix.csv · ranking.csvraw and normalized scores
  • bgc/antiSMASH_regions/biosynthetic clusters per strain
  • raw/*.fastq.gz + checksumsraw reads, returned to you
  • code/ (Nextflow + R, Docker)to reproduce every number

How the scores are built

Each genome is assembled, quality-checked and annotated. Plant-growth and biocontrol genes are grouped into functional blocks (PLaBAse-type annotation), normalized across strains and compared with commercial benchmarks.

Crop and goal set the weight of each trait. Scores become probabilities with a calibration model, and consortia are scored on combined coverage, minus predicted antagonism.

Probabilities rank candidates before costly trials. They do not replace greenhouse and field validation, which is the next step this report plans.