Low concentration cell painting images enable the identification of highly potent compounds

Image-based models that use features extracted from cell microscopy images can estimate the activity of small molecules in various biological assays. Typically, models are trained on images stained by an optimized protocol (e.g. Cell Painting) after exposure to a fairly high small molecule concentration (referred to as 'image concentration') of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mn>10</mml:mn> <mml:mspace></mml:mspace> <mml:mi>μ</mml:mi> <mml:mtext>M</mml:mtext></mml:mrow> </mml:math> or higher. Low concentration images (e.g. <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mn>0.16</mml:mn></mml:mrow> </mml:math> μM, <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mn>0.8</mml:mn></mml:mrow> </mml:math> μM, <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mn>4</mml:mn></mml:mrow> </mml:math> μM) tend to yield models with worse performance. In this work, we nevertheless report a practical use for low image concentration data. We propose the combination of well-performing models trained at higher image concentrations, with lower image concentration for inference to identify more potent compounds. We show that this approach improves on the conventional method (directly training a high-potency model) in 65 <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mo>%</mml:mo></mml:math> of assays investigated in terms of AUC-ROC, and 75 <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mo>%</mml:mo></mml:math> of assays in terms of RIPtoP-corrected AUC-PR.