Skip to content

yqia182/iterative-calculation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

63 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Introduction

Preview

Predicted and observed datasets

Gompertz Modeling of Growth Curves

There is a modification to the Gompertz model so as to fit the model in to microorganism growth. If the population of the microorganism at a certain time is translated into optical density, then there is (Roncoroni, M., 2014),

Modified Gompertz Model
minOD - Minimum Optical Density
maxOD - Maximum Optical Density (Carrying Capacity)
lag - The Duration of Lag Phase of the Growth
μmax - Maximum Growth Rate (at Exponential Growth Phase)

Brute-Force Search & Greedy Algorithm

To calculate the four parameters above, assumpted values for these four are tried out.
Each combination of four values will form a formula. For each formula, the coeffcient of determination (coef) can be calcalated out with the dataset from observations.
The coef is used to evaluate whether the four values are a good fit into your observed dataset.

Time and space complexity of Brute-Force Search algorithm would be too high for this calculation problem. In other words, there are way too many values to try out for these four parameters. Since this problem has greedy-choice properties, an alternative method, expectation-maximization (EM) algorithm is implemented in this project.

Citations

Roncoroni, M. (2014). Quantitative trait loci mapping in winemaking yeast (Doctoral dissertation, ResearchSpace@ Auckland).

Instructions

Configuration

Settings

Database

Install Mysql to your PC if you do not yet have a Mysql database. Configure the database name and credentials as, for example,

spring.datasource.url=jdbc:mysql://127.0.0.1:3306/iterative?allowMultiQueries=true&useUnicode=true&characterEncoding=UTF-8&useSSL=false&serverTimezone=Asia/Shanghai&allowPublicKeyRetrieval=true
spring.datasource.username=iterative
spring.datasource.password=123456

Parameters

Predict a range for each parameter in application-dev.properties. For example,

iterative.param.suggestion.lag-range-min = 400
iterative.param.suggestion.lag-range-max = 1200
iterative.param.suggestion.rate-range-min =0.00200
iterative.param.suggestion.rate-range-max = 0.00800
iterative.param.suggestion.minOD-range-min= 0.00
iterative.param.suggestion.minOD-range-max= 0.080
iterative.param.suggestion.maxOD-range-min=0.90
iterative.param.suggestion.maxOD-range-max=1.70

This does not have to be precise. The ranges should be wide enough for the actual values falling into.

Dataset File

The dataset must be arranged in a CSV file like, Csv File Demo

Configure the file path and name. For example,

iterative.filepath.observed-data-csv-file-path=D:\Desktop\1.csv

URI

Navigate to locahost:8080 and click on "Run All Samples" or "Run One Sample".

Results

Rrefer to the table named final_result_permanent_entity and select the results with hightest coef. Always export the results before you drop any table.

About

Greedy Algorithm Implemented for Gompertz Modeling

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published