We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Microplastic-LOSO v1.0.0: code and results for 'Recognizing the study is not predicting adsorption: leave-one-study-out re-evaluation of machine-learning models for microplastic removal by biomass composites'
AI summary Read the abstract
This is a code and data archive, not a new experiment. It re-checks earlier machine-learning models that predicted how well biomass materials soak up microplastics, testing them by leaving out each of 24 original studies one at a time to see how well the models actually generalize.
Code, result files and figures for a leave-one-study-out re-evaluation of machine-learning models of microplastic adsorption on biomass composites (Li et al., Journal of Hazardous Materials 2026, doi:10.1016/j.jhazmat.2025.140700). A cell-level parser reads the data table of the source supplementary information; random forest, gradient boosting and XGBoost are evaluated for the equilibrium capacity (qe) and equilibrium time (t) on random 80/20 splits and with each of the 24 source studies held out, together with a study-mean predictor that uses no input, bootstrap intervals, a k-nearest-neighbour applicability domain and a sensitivity analysis for a probable transcription error. Contents: analysis scripts (software/src) with a vendored copy of the small nanodiscovery_core library, pinned dependencies, the parsing report and all result files, figures (PNG 600 dpi and PDF) and the graphical abstract, the supplementary tables as JSON, data and model cards, and a SHA-256 manifest. The source supplementary file has no licence and is not redistributed; the archive contains no record-level values from it, and the README gives the download command and SHA-256 hash.
More Papers Like This
Predicting aqueous sorption of organic pollutants on microplastics with machine learning
AI summary Read the abstract
Researchers developed machine learning models to predict how organic pollutants bind to microplastics in water, using data from 475 published experiments. The models outperformed traditional approaches by accounting for properties of both the microplastics and the pollutants simultaneously. The study provides a more universal tool for understanding how microplastics can transport and concentrate harmful chemicals in freshwater systems.
Elucidating microplastic adsorption mechanisms in biomass composite materials through interpretable machine learning
AI summary Read the abstract
Researchers used interpretable machine learning to study how biomass composite materials adsorb microplastics from water. They found that initial microplastic concentration and surface electrical potential were the most important factors determining adsorption effectiveness. The study demonstrates that data-driven approaches can help design more efficient and sustainable materials for removing microplastics from contaminated water.
Machine Learning to Predict the Adsorption Capacity of Microplastics
AI summary Read the abstract
Researchers developed machine learning models to predict the adsorption capacity of microplastics for chemical pollutants, providing a computational tool to better understand how microplastics act as vectors for contaminant dispersal in aquatic environments.
Versatile in silico modelling of microplastics adsorption capacity in aqueous environment based on molecular descriptor and machine learning
AI summary Read the abstract
Researchers developed machine learning models using molecular descriptors to predict the adsorption capacity of microplastics for organic pollutants in aqueous environments, achieving high accuracy across multiple polymer types and enabling faster environmental risk assessment.
Reproducible Data and R Code for a Meta-analysis and Machine-learning Study of Aged Micro- and Nanoplastic Phytotoxicity
AI summary Read the abstract
Scientists combined data from 51 studies to see how "weathered" microplastics, the tiny plastic bits broken down by sun and weather, affect plant health differently than fresh plastic. This matters because microplastics are entering our food supply through crops, and understanding how environmental aging changes their toxicity helps researchers better predict risks to the food we eat.
Research digests by email
When a large batch of papers lands in the Atlas, we read through it and send a short write-up of what stood out.