Analyzing Patient Trajectories With Artificial Intelligence

In digital medicine, patient data typically record health events over time (eg, through electronic health records, wearables, or other sensing technologies) and thus form unique patient trajectories. Patient trajectories are highly predictive of the future course of diseases and therefore facilitate effective care. However, digital medicine often uses only limited patient data, consisting of health events from only a single or small number of time points while ignoring additional information encoded in patient trajectories. To analyze such rich longitudinal data, new artificial intelligence (AI) solutions are needed. In this paper, we provide an overview of the recent efforts to develop trajectory-aware AI solutions and provide suggestions for future directions. Specifically, we examine the implications for developing disease models from patient trajectories along the typical workflow in AI: problem definition, data processing, modeling, evaluation, and interpretation. We conclude with a discussion of how such AI solutions will allow the field to build robust models for personalized risk scoring, subtyping, and disease pathway discovery.

MicroRNA-directed pathway discovery elucidates an miR-221/222–mediated regulatory circuit in class switch recombination

MicroRNAs (miRNAs, miRs) regulate cell fate decisions by post-transcriptionally tuning networks of mRNA targets. We used miRNA-directed pathway discovery to reveal a regulatory circuit that influences Ig class switch recombination (CSR). We developed a system to deplete mature, activated B cells of miRNAs, and performed a rescue screen that identified the miR-221/222 family as a positive regulator of CSR. Endogenous miR-221/222 regulated B cell CSR to IgE and IgG1 in vitro, and miR-221/222–deficient mice exhibited defective IgE production in allergic airway challenge and polyclonal B cell activation models in vivo. We combined comparative Ago2-HITS-CLIP and gene expression analyses to identify mRNAs bound and regulated by miR-221/222 in primary B cells. Interrogation of these putative direct targets uncovered functionally relevant downstream genes. Genetic depletion or pharmacological inhibition of Foxp1 and Arid1a confirmed their roles as key modulators of CSR to IgE and IgG1.

Diet-induced modifications to human microbiome reshape colonic homeostasis in irritable bowel syndrome

Changes in microbiome composition have been associated with a wide array of human diseases, turning the human microbiota into an attractive target for therapeutic intervention. Yet clinical translation of these findings requires the establishment of causative connections between specific microbial taxa and their functional impact on host tissues. Here, we colonized gut organ cultures with longitudinal microbiota samples collected from newly-diagnosed and therapy-naive irritable bowel syndrome (IBS) patients under low-FODMAP (fermentable Oligo-, Di-, Mono-saccharides and Polyols) diet. We show that post-diet microbiota regulates intestinal expression of inflammatory and neuro-muscular gene-sets. Specifically, we identify Bifidobacterium adolescentis as a diet-sensitive pathobiont that alters tight junction integrity and disrupts gut barrier functions. Collectively, we present a unique pathway discovery approach for mechanistic dissection and identification of functional diet-host-microbiota modules. Our data support the hypothesis that the gut microbiota mediates the beneficial effects of low-FODMAP diet, and reinforce the potential feasibility of microbiome based-therapies in IBS.

Best Paper Selection

Fabregat A, Magret M, Ferré JA, Vernet A, Guasch N, Rodríguez A, Gómez J, Bodí M. A Machine Learning decision-making tool for extubation in Intensive Care Unit patients. https://www.sciencedirect.com/science/article/abs/pii/S0169260720317028?via%3Dihub Kempa-Liehr AW, Lin CYC, Britten R, Armstrong D, Wallace J, Mordaunt D, O’Sullivan M. Healthcare pathway discovery and probabilistic machine learning. https://www.sciencedirect.com/science/article/abs/pii/S1386505619308068?via%3Dihub Li Y, Nair P, Lu XH, Wen Z, Wang Y, Dehaghi AAK, Miao Y, Liu W, Ordog T, Biernacka JM, Ryu E, Olson JE, Frye MA, Liu A, Guo L, Marelli A, Ahuja Y, Davila-Velderrain J, Kellis M. Inferring multimodal latent topics from electronic health records. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7242436/ Weemaes M, Martens S, Cuypers L, van Elslande J, Hoet K, Welkenhuysen J, Goossens R, Wouters S, Houben E, Jeuris E, Jeuris K, Laenen L, Bruyninckx K, Beuselinck K, André E, Depypere M, Desmet S, Lagrou K, Van Ranst M, Verdonck AKLC, Goveia J. Laboratory information system requirements to manage the COVID-19 pandemic: A report from the Belgian national reference testing center. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7197526/

Analyzing Patient Trajectories With Artificial Intelligence (Preprint)

UNSTRUCTURED In digital medicine, patient data typically record health events over time (eg, through electronic health records, wearables, or other sensing technologies) and thus form unique patient trajectories. Patient trajectories are highly predictive of the future course of diseases and therefore facilitate effective care. However, digital medicine often uses only limited patient data, consisting of health events from only a single or small number of time points while ignoring additional information encoded in patient trajectories. To analyze such rich longitudinal data, new artificial intelligence (AI) solutions are needed. In this paper, we provide an overview of the recent efforts to develop trajectory-aware AI solutions and provide suggestions for future directions. Specifically, we examine the implications for developing disease models from patient trajectories along the typical workflow in AI: problem definition, data processing, modeling, evaluation, and interpretation. We conclude with a discussion of how such AI solutions will allow the field to build robust models for personalized risk scoring, subtyping, and disease pathway discovery.

Biosynthesis and synthetic biology of psychoactive natural products

The biosynthetic logic employed by Nature in the construction of psychoactive natural products is reviewed, in addition to biological activities, methodologies enabling pathway discovery, and engineering applications.