Source code and modeling dataset for 'Optimizing lipid substrate composition reveals contributions of β-hydroxybutyrate and palmitic acid to milk lipid synthesis'

Published: 29 October 2026| Version 1 | DOI: 10.17632/kp9592gbwt.1
Contributors:
,
,
,
,
,
,
,
,

Description

This dataset contains the source code, modeling dataset, and supporting documentation for the study entitled "Optimizing lipid substrate composition reveals contributions of β-hydroxybutyrate and palmitic acid to milk lipid synthesis". The files support the response surface methodology (RSM) and machine learning (ML) analyses used to model and optimize lipid substrate combinations for triglyceride production in bovine mammary epithelial cells. - README_ML.docx provides the software requirements, input data specifications, model implementation details, execution instructions, and information required to reproduce the ML analyses. - ML_Source_Code.docx contains the complete Python source code used for ML model comparison, nested grouped cross-validation, hyperparameter tuning, final model refitting, constrained numerical optimization, and figure generation. - ML_Modeling_Dataset.xlsx contains the experimental dataset used for RSM and ML analyses. The dataset includes the concentrations of acetate (ACA), β-hydroxybutyrate (BHBA), palmitic acid (PA), and stearic acid (SA), together with the corresponding measured triglyceride (TG) concentrations. - Supplementary Code 1.docx contains the complete Python source code used for RSM fitting, model-adequacy assessment, response-surface prediction, and constrained numerical optimization.

Files

Institutions

Categories

Cell Biology, Animal Biotechnology, Machine Learning

Funders

Licence