Icon

bacterium_​with_​MC_​training

The workflow has two types of input files:
1) Several csv input files read in a loop. These files contain the measurement of individual bacteria for twenty consecutive time points. The images contained two fluorescent channels, green and magenta.
2) Excel file is loaded which contains, for each channel, the frame at which the first microcompartment (MC) appears.
Both inputs are joined according to the file name. Then the data is split to magenta and green channel data. For each channel, a random forest learner is trained on of the data, taking into account four intensity based measurements and the identified first frame of MC appearance. The models are saved and can be used to classify data of the same type as input 1. The results are pivoted and joined to export the time frame of first appearance of a MC per channel and input file.

Input 1: Read series of input files containing the time series measurements for a cropped single bacterium each. Input 2: Read "Sheet1" of a single excel file listing the frame index of the first appearance of amicrocompartment for each channel for all of the files in the series of single bacteriummeasurements. Pre-processing: Joining measurements and information on time frameindex of first appearance of a microcompartment per channel. Training: Partition input data and train randomforest learner for magenta signal. Training: Partition input data and train randomforest learner for green signal. Result generation: Use trained modelsto generate classifications for magentaand green signals for the full dataset. Downstream processing: Join data and generate frame of first MCappearance for green and magenta channel for each of the input files. Node 11Node 12read individual bacteria time series measurementsNode 14add classificationfor magenta signaldepending on timeframe index in input 2add classificationfor green signaldepending on timeframe index in input 2extract file name from pathNode 20read indeces for MC genesisNode 24Node 26combine information of input files depending on file nameNode 29split magentaand green channel information Node 31magentagreenNode 34Node 35Node 36Node 37Node 38Node 39duplicate somecolumns so theywill be present inboth split setsNode 41Node 42Node 43read magenta modelNode 48Node 49read green modelNode 51Node 52first frame green MCfirst frame magenta MCNode 57time index of first MC classification per fileNode 59Node 60Node 61 List Files/Folders Table Row ToVariable Loop Start CSV Reader Loop End Rule Engine Rule Engine Column Expressions Path to String Excel Reader Sorter Column Expressions Joiner Sorter Column Splitter Column Filter Partitioning Partitioning Random ForestLearner Random ForestPredictor Scorer Random ForestPredictor Scorer Random ForestLearner Column Appender Column Filter Model Writer Model Writer Model Reader Random ForestPredictor Random ForestPredictor Model Reader Joiner Column Resorter Pivoting Pivoting Joiner Column Rename Scorer Scorer CSV Writer Input 1: Read series of input files containing the time series measurements for a cropped single bacterium each. Input 2: Read "Sheet1" of a single excel file listing the frame index of the first appearance of amicrocompartment for each channel for all of the files in the series of single bacteriummeasurements. Pre-processing: Joining measurements and information on time frameindex of first appearance of a microcompartment per channel. Training: Partition input data and train randomforest learner for magenta signal. Training: Partition input data and train randomforest learner for green signal. Result generation: Use trained modelsto generate classifications for magentaand green signals for the full dataset. Downstream processing: Join data and generate frame of first MCappearance for green and magenta channel for each of the input files. Node 11Node 12read individual bacteria time series measurementsNode 14add classificationfor magenta signaldepending on timeframe index in input 2add classificationfor green signaldepending on timeframe index in input 2extract file name from pathNode 20read indeces for MC genesisNode 24Node 26combine information of input files depending on file nameNode 29split magentaand green channel information Node 31magentagreenNode 34Node 35Node 36Node 37Node 38Node 39duplicate somecolumns so theywill be present inboth split setsNode 41Node 42Node 43read magenta modelNode 48Node 49read green modelNode 51Node 52first frame green MCfirst frame magenta MCNode 57time index of first MC classification per fileNode 59Node 60Node 61 List Files/Folders Table Row ToVariable Loop Start CSV Reader Loop End Rule Engine Rule Engine Column Expressions Path to String Excel Reader Sorter Column Expressions Joiner Sorter Column Splitter Column Filter Partitioning Partitioning Random ForestLearner Random ForestPredictor Scorer Random ForestPredictor Scorer Random ForestLearner Column Appender Column Filter Model Writer Model Writer Model Reader Random ForestPredictor Random ForestPredictor Model Reader Joiner Column Resorter Pivoting Pivoting Joiner Column Rename Scorer Scorer CSV Writer

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: mheldb
Created at: 2024-12-10
On NodePit since: 2024-12-15
Last update: 2026-01-23
Created with KNIME version: v4.7.8

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!