Fig. 4: Novel candidate immunotherapy drug targets. | Nature Machine Intelligence

Fig. 4: Novel candidate immunotherapy drug targets.

From: Immunotherapy drug target identification using machine learning and patient-derived tumour explant validation

Fig. 4: Novel candidate immunotherapy drug targets.The alternative text for this image may have been generated using AI.

a, OSM and OSMR are novel candidate immunotherapy drug targets. For comparison purposes, the predicted probability for PTPN22 (a previously suggested preclinical target), and a density plot showing the predictions for known immuno-oncology targets and non-targets are provided. b, Pathway-based permutation importance for PTPN22. Colour bar indicates the importance of manually curated pathways for predicting PTPN22 as a drug target. APP, antigen processing and presentation. c, Pathway-based permutation importance for the OSM–OSMR axis. Colour bar indicates the importance of manually curated pathways for predicting the OSM–OSMR axis as an immunotherapy candidate. In b and c, the grey pathways are from the Reactome pathway hierarchy and were not perturbed either because the target does not belong to that pathway or because there were insufficient edges to permute.

Source data

Back to article page