Coverage Score: A Model Agnostic Method to Efficiently Explore Chemical Space.

Selecting the most appropriate compounds to synthesize and test is a vital aspect of drug discovery. Methods like clustering and diversity present weaknesses in selecting the optimal sets for information gain. Active learning techniques often rely on an initial model and computationally expensive semi-supervised batch selection. Herein, we describe a new subset-based selection method, Coverage Score, that combines Bayesian statistics and information entropy to balance representation and diversity to select a maximally informative subset. Coverage Score can be influenced by prior selections and desirable properties. In this paper, subsets selected through Coverage Score are compared against subsets selected through model-independent and model-dependent techniques for several datasets. In drug-like chemical space, Coverage Score consistently selects subsets that lead to more accurate predictions compared to other selection methods. Subsets selected through Coverage Score produced Random Forest models that have a root-mean-square-error up to 12.8% lower than subsets selected at random and can retain up to 99% of the structural dissimilarity of a diversity selection.

Abnormal myocardial blood flow in children with mild/moderate aortic stenosis.

OBJECTIVE: To quantify myocardial blood flow in infants and children with mild or moderate aortic stenosis using adenosine-infusion cardiac magnetic resonance. BACKGROUND: It is unclear whether asymptomatic children with mild/moderate aortic stenosis have myocardial abnormalities. In addition, cardiac magnetic resonance-determined normative myocardial blood flow data in children have not been reported. METHODS: We studied 31 infants and children with either haemodynamically normal hearts (n=20, controls) or mild/moderate aortic stenosis (n=11). The left ventricular myocardium was divided into six segments, and the change in average segmental signal intensity during contrast transit was used to quantify absolute flow (ml/g/minute) at rest and during adenosine infusion by deconvolution of the tissue curves with the arterial input of contrast. RESULTS: In all the cases, adenosine was well tolerated without complications. The mean pressure gradient between the left ventricle and the ascending aorta was higher in the aortic stenosis group compared with controls (24 versus 3 mmHg, p<0.001). Left ventricular wall mass was slightly higher in the aortic stenosis group compared with controls (65 versus 50 g/m², p<0.05). After adenosine treatment, both the absolute increase in myocardial blood flow (p<0.0001) and the hyperaemic flow significantly decreased (p<0.001) in children with mild/moderate aortic stenosis compared with controls. CONCLUSION: Abnormal myocardial blood flow in children with mild/moderate aortic stenosis may be an important therapeutic target.