publiccode.yml

publiccodeYmlVersion: 0.5.0
name: Plasmopy
url: https://github.com/agroscope-ch/plasmopy
landingURL: https://github.com/agroscope-ch/plasmopy/blob/main/README.md
softwareVersion: "1.0"
releaseDate: 2025-07-09
platforms:
  - linux
categories:
  - other
fundedBy:
  - name: Agroscope
    uri: https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&uact=8&ved=2ahUKEwjxy5z2_cGSAxUW0AIHHbn7BXMQFnoECCIQAQ&url=https%3A%2F%2Fwww.agroscope.admin.ch%2Fagroscope%2Fen%2Fhome.html&usg=AOvVaw38MIoPtH6NosgSpoaWsqSU&opi=89978449
usedBy:
  - Agroscope
developmentStatus: stable
softwareType: standalone/desktop
organisation:
  uri: https://ld.admin.ch/ou/10003634
  name: Agroscope
description:
  en:
    localisedName: Plasmopy
    shortDescription: Plasmopara viticola (downy mildew) infection risk prediction
      model for grapevine.
    longDescription: >-
      Infection prediction modeling of Plasmopara viticola life cycle stages in
      vineyards using weather data. This project represents a revisited and
      customizable Python version of the VitiMeteo-Plasmopara model in use at
      Agroscope (Switzerland) for downy mildew infection forecasting.


      Input data - Weather variables


      Input data needs to be in the form of timeseries in a semicolon-delimited
      CSV file, where the positioning of columns must be precisely kept in the
      following order:


      1) Date and time of measurement in the %d.%m.%Y %H:%M local timezone
      format.

      2) Average temperature [°C] in degrees Celsius.

      3) Relative humidity [%] in percentages.

      4) Rainfall intensity [mm/h] in millimeters per hour.

      5) Leaf wetness [min] in minutes of wetness over 10 minutes (by default,
      or whatever sampling period is available).


      Such timeseries data is publicly available for Switzerland from the
      Agroscope weather stations, accessible at Agrometeo.


      Examples of input weather data are available in data/input/.


      Be sure to update accordingly the most important input-dependent
      parameters in config/main.yaml: measurement_time_interval,
      computational_time_steps, algorithmic_time_steps.


      The file can be managed directly, or through the streamlit web-app, as
      detailed in the README.


      Optional input data - Spore counts

      Input spore counts data (e.g. from qPCR, microscopy, ...) columns must be
      ordered and formatted in the following way:

      1) Date of daily spore counts in the %d.%m.%Y %H:%M local timezone format.

      2) Spore counts.


      Infection model

      The main.py and infection_model.py Python scripts orchestrate a
      computational framework for modeling infection dynamics based on
      meteorological and biological parameters. Below, the main components of
      the methodology are outlined:


      Data Preprocessing and Initialization

      The script requires input data formatted as a time-series DataFrame, model
      parameters provided via a configuration file, and supporting metadata such
      as time zones and column formats. Model parameters are validated and
      logged, ensuring consistency across computational stages.


      Determination of Oospore Maturation

      The get_oospore_maturation_date function identifies the date of oospore
      maturation using daily temperature data. If a predefined maturation date
      is unavailable, the function computes it based on degree-day thresholds
      and writes results to a log file. Missing data or unmet threshold
      conditions halt the process.


      Infection Event Construction

      The function get_infection_events_dictionary compiles and organizes data
      into a dictionary containing key infection event datetimes (e.g., oospore
      germination, sporulation) and their associated parameters (e.g., sporangia
      density, spore lifespan).


      Primary Infection Prediction


      The run_infection_model function coordinates the infection prediction
      process, starting with primary infection stages:


      Oospore Germination: Simulated based on relative humidity, temperature
      thresholds, and environmental conditions.


      Oospore Dispersion: Modeled as a rainfall-driven process with latency
      considerations.


      Oospore Infection: Dependent on leaf wetness, degree-hour accumulation,
      and environmental conditions.


      Secondary Infection and Incubation Dynamics


      Subsequent infection stages include:


      Incubation Period: Duration is computed based on mean temperatures.


      Sporulation: Triggered by environmental factors such as humidity,
      temperature, and darkness duration.


      Sporangia Density: Simulated as a function of temperature and latency
      parameters.


      Spore Lifespan: Calculated using vapor pressure and lifespan constants.


      Secondary Infections: Modeled based on spore availability, temperature
      thresholds, and leaf wetness conditions.


      Logging and Error Handling

      The script employs robust error logging to identify issues such as missing
      configuration parameters or unfulfilled biological conditions. Warnings
      are issued if model requirements (e.g., maturation thresholds) are unmet.


      Output

      The final output is a structured pickle dictionary summarizing infection
      event properties and datetimes. This dictionary serves as the primary
      interface for downstream analyses or reporting. 

      Additional output files included in data/output/:

      PDF and HTML graphs of the simulated infection events across the input
      datetime range;

      a logfile detailing the simulation data pre-processing, model parameters
      and eventual workflow errors;

      the post-processing weather variables input file;

      a summary description of the successful primary and secondary infection
      events.

      The output files as well as the model parameters and input files for
      running new simulations can be interactively explored and configured
      through the streamlit web-app, activated with the command:


      make app
    documentation: https://github.com/agroscope-ch/plasmopy/blob/main/README.md
    apiDocumentation: https://github.com/agroscope-ch/plasmopy/blob/main/MANUAL.md
    features:
      - Predict Plasmopara viticola life cycle infection stage datetimes based
        on weather data.
      - Plot Plasmopara viticola life cycle infection stages timeline.
      - Print predicted Plasmopara viticola infection stage datetimes datasets
        for further analysis.
      - Modify and configure all Plasmopara viticola life cycle parameters and
        climatic infection thresholds and algorithms.
      - Add observed field spore counts data to the model and adapt the
        infection risk computation consequently.
      - Run all configuration and model simulations through the dedicated
        interactive Python-Streamlit web-app. Readily deployable on web-servers.
    screenshots:
      - https://github.com/agroscope-ch/plasmopy/blob/main/images/plasmopy_screen1.jpg
      - https://github.com/agroscope-ch/plasmopy/blob/main/images/plasmopy_screen3.jpg
legal:
  license: AGPL-3.0-only
maintenance:
  type: internal
  contacts:
    - name: Livio Ruzzante
      email: livio.ruzzante@agroscope.admin.ch
      affiliation: Agroscope
localisation:
  localisationReady: true
  availableLanguages:
    - en