Gasser, L. | Search bioRxiv

bioRxiv2023

ADORE: A benchmark dataset for machine learning in ecotoxicology

The use of machine learning for predicting ecotoxicological outcomes is promising, but underutilized. The curation of data with informative features requires both expertise in machine learning as well as a strong biological and ecotoxicological background, which we consider a barrier of entry for this kind of research. Additionally, model performances can only be compared across studies when the same dataset, cleaning, and splittings were used. Therefore, we provide ADORE, an extensive and well-described dataset on acute aquatic toxicity in three relevant taxonomic groups (fish, crustaceans, and algae). The core dataset describes ecotoxicological experiments and is expanded with phylogenetic and species-specific data on the species as well as chemical properties and molecular representations. Apart from challenging other researchers to try and achieve the best model performances across the whole dataset, we propose specific relevant challenges on subsets of the data and include datasets and splittings corresponding to each of these challenge as well as in-depth characterization and discussion of train-test splitting approaches. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=104 SRC="FIGDIR/small/542160v2_ufig1.gif" ALT="Figure 1"> View larger version (25K): org.highwire.dtl.DTLVardef@1a58fc3org.highwire.dtl.DTLVardef@817f57org.highwire.dtl.DTLVardef@1cf5626org.highwire.dtl.DTLVardef@120e814_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioinformatics↗

bioRxiv2022

Retention time prediction to facilitate molecularstructure identification with tandem mass spectrometry

Comparing measured and predicted chromatographic retention time can improve molecular structure assignment in applications such as coupled liquid chromatography-tandem mass spectrometry. We assess a range of different machine learning methods to predict hydrophobicity, a molecular property that can be used as a proxy for retention time. The performance of the models is evaluated on the benchmark Martel and SAMPL7 datasets. We find that more powerful models perform better when predicting in-sample but not necessarily when generalizing to out-of-sample molecular families. We also find that ensemble methods can outperform individual models. Additionally, a multitask learning model shows promise for improving the generalization ability of graph neural networks for hydrophobicity prediction. Finally, we discuss how the ability of graph neural networks to generalize for molecular property prediction could be improved further.

bioinformatics↗