Deep learning models capture histological disease activity in Crohn's Disease and Ulcerative Colitis with high fidelity.

BACKGROUND AND AIMS: Histologic disease activity in Inflammatory Bowel Disease (IBD) is associated with clinical outcomes and is an important endpoint in drug development. We developed deep learning models for automating histological assessments in IBD. METHODS: Histology images of intestinal mucosa from phase 2 and phase 3 clinical trials in Crohn's disease (CD) and Ulcerative Colitis (UC) were used to train artificial intelligence (AI) models to predict the Global Histology Activity Score (GHAS) for CD and Geboes histopathology score for UC. Three AI methods were compared. AI models were evaluated on held-back testing sets and model predictions were compared against an expert central reader and five independent pathologists. RESULTS: The model based on multiple instance learning and the attention mechanism (SA-AbMILP) demonstrated the best performance among competing models. AI modeled GHAS and Geboes sub-grades matched central readings with moderate to substantial agreement, with accuracies ranging from 65% to 89%. Furthermore, the model was able to distinguish the presence and absence of pathology across four selected histological features with accuracies for colon, in both CD and UC, ranging from 87% to 94% and, for CD ileum, ranging from 76% to 83%. For both CD and UC, and across anatomical compartments (ileum and colon) in CD, comparable accuracies against central readings were found between the model assigned scores and scores by an independent set of pathologists. CONCLUSIONS: Deep learning models based upon GHAS and Geboes scoring systems were effective at distinguishing between the presence and absence of IBD microscopic disease activity.

Community Assessment of the Predictability of Cancer Protein and Phosphoprotein Levels from Genomics and Transcriptomics.

Cancer is driven by genomic alterations, but the processes causing this disease are largely performed by proteins. However, proteins are harder and more expensive to measure than genes and transcripts. To catalyze developments of methods to infer protein levels from other omics measurements, we leveraged crowdsourcing via the NCI-CPTAC DREAM proteogenomic challenge. We asked for methods to predict protein and phosphorylation levels from genomic and transcriptomic data in cancer patients. The best performance was achieved by an ensemble of models, including as predictors transcript level of the corresponding genes, interaction between genes, conservation across tumor types, and phosphosite proximity for phosphorylation prediction. Proteins from metabolic pathways and complexes were the best and worst predicted, respectively. The performance of even the best-performing model was modest, suggesting that many proteins are strongly regulated through translational control and degradation. Our results set a reference for the limitations of computational inference in proteogenomics. A record of this paper's transparent peer review process is included in the Supplemental Information.

Mol-CycleGAN: a generative model for molecular optimization.

Designing a molecule with desired properties is one of the biggest challenges in drug development, as it requires optimization of chemical compound structures with respect to many complex properties. To improve the compound design process, we introduce Mol-CycleGAN-a CycleGAN-based model that generates optimized compounds with high structural similarity to the original ones. Namely, given a molecule our model generates a structurally similar one with an optimized value of the considered property. We evaluate the performance of the model on selected optimization objectives related to structural properties (presence of halogen groups, number of aromatic rings) and to a physicochemical property (penalized logP). In the task of optimization of penalized logP of drug-like molecules our model significantly outperforms previous results.