Calder, P. | Search bioRxiv

bioRxiv2025

Quantification of esterified oxylipins following HILIC-fractionation of lipid classes

Several oxylipins are lipid mediators derived from the oxidation of polyunsaturated fatty acids (PUFAs). The majority of oxylipins in biological samples occurs esterified in neutral lipids (nLs) and phospholipids (PLs). They are commonly quantified indirectly following alkaline hydrolysis providing excellent sensitivity but the information in which lipid classes the oxylipins occurred in is lost. The direct analysis of oxidized lipids is currently not sensitive enough to detect all esterified oxylipins. Here, a new hydrophilic interaction liquid chromatography (HILIC) based lipid class fractionation using solid-phase extraction (SPE) cartridges was developed separating lipids into nLs and 4 PL fractions using a single column. Esterified oxylipins in the fractions were quantified following alkaline hydrolysis to sensitively pinpoint in which lipid classes they are bound in plasma. The fractionation was extensively characterized for different lipid extracts demonstrating high separation efficiency and recovery using labeled standards and untargeted analysis of endogenous lipids. Esterified oxylipins in the fractions were quantitatively detected. Based on the results from two independent human plasma pools including SRM1950 it is shown that: hydroxy-linoleic acid- and hydroxy--linolenic acid-derived oxylipins are preferably bound to nLs whereas long chain hydroxy-PUFAs and PUFAs (i.e. ARA EPA and DHA) are predominantly esterified to phospholipid classes. Supplementation of n3-PUFAs for 12 months led to an increase in EPA- and -DHA-derived oxylipins in all lipid fractions with the highest increase of hydroxy-PUFAs in nLs. This demonstrates a precursor PUFA-dependent binding of oxylipins and a direct effect of diet on esterified oxylipins in plasma.

biochemistry↗

bioRxiv2023

A Bayesian active learning platform for scalable combination drug screens

Large-scale combination drug screens are generally considered intractable due to the immense number of possible combinations. Existing approaches use ad hoc fixed experimental designs then train machine learning models to impute novel combinations. Here we propose BATCHIE, an orthogonal approach that conducts experiments dynamically in batches. BATCHIE uses information theory and probabilistic modeling to design each batch to be maximally informative based on the results of previous experiments. On retrospective experiments from previous large-scale screens, BATCHIE designs rapidly discover highly effective and synergistic combinations. To validate BATCHIE prospectively, we conducted a combination screen on a collection of pediatric cancer cell lines using a 206 drug library. After exploring only 4% of the 1.4M possible experiments, the BATCHIE model was highly accurate at predicting novel combinations and detecting synergies. Further, the model identified a panel of top combinations for Ewing sarcomas, all of which were experimentally confirmed to be effective, including the rational and translatable top hit of PARP plus topoisomerase I inhibition. These results demonstrate that adaptive experiments can enable large-scale unbiased combination drug screens with a relatively small number of experiments, thereby powering a new wave of combination drug discoveries. BATCHIE is open source and publicly available (https://github.com/tansey-lab/batchie).

bioinformatics↗