Spurred by advances in processing power, memory, storage, and an unprecedented wealth of data, computers are being asked to tackle increasingly complex learning tasks, often with astonishing success. Computers have now mastered a popular variant of poker, learned the laws of physics from experimental data, and become experts in video games – tasks which would have been deemed impossible not too long ago. In parallel, the number of companies centered on applying complex data analysis to varying industries has exploded, and it is thus unsurprising that some analytic companies are turning attention to problems in healthcare. The purpose of this review is to explore what problems in medicine might benefit from such learning approaches and use examples from the literature to introduce basic concepts in machine learning. It is important to note that seemingly large enough medical data sets and adequate learning algorithms have been available for many decades – and yet, although there are thousands of papers applying machine learning algorithms to medical data, very few have contributed meaningfully to clinical care. This lack of impact stands in stark contrast to the enormous relevance of machine learning to many other industries. Thus part of my effort will be to identify what obstacles there may be to changing the practice of medicine through statistical learning approaches, and discuss how these might be overcome.

Introduction Heart failure with preserved ejection fraction (HFpEF) is a heterogeneous clinical syndrome in need of improved phenotypic classification. We sought to evaluate whether unbiased clustering analysis using dense phenotypic data (“phenomapping”) could identify phenotypically distinct HFpEF categories. Methods and Results We prospectively studied 397 HFpEF patients and performed detailed clinical, laboratory, electrocardiographic, and echocardiographic phenotyping of the study participants. We used several statistical learning algorithms, including unbiased hierarchical cluster analysis of phenotypic data (67 continuous variables) and penalized model-based clustering to define and characterize mutually exclusive groups comprising a novel classification of HFpEF. All phenomapping analyses were performed blinded to clinical outcomes, and Cox regression was used to demonstrate the clinical validity of phenomapping. The mean age was 65±12 years, 62% were female, 39% were African-American, and comorbidities were common. Although all patients met published criteria for the diagnosis of HFpEF, phenomapping analysis classified study participants into 3 distinct groups that differed markedly in clinical characteristics, cardiac structure/function, invasive hemodynamics, and outcomes (e.g., pheno-group #3 had an increased risk of HF hospitalization [hazard ratio 4.2, 95% CI 2.0–9.1] even after adjustment for traditional risk factors [P<0.001]). The HFpEF pheno-group classification, including its ability to stratify risk, was successfully replicated in a prospective validation cohort (n=107). Conclusions Phenomapping results in novel classification of HFpEF. Statistical learning algorithms, applied to dense phenotypic data, may allow for improved classification of heterogeneous clinical syndromes, with the ultimate goal of defining therapeutically homogeneous patient subclasses.

SUMMARY In vertebrates, activation of innate immunity is an early response to injury, implicating it in the regenerative process. However, the mechanisms by which innate signals might regulate stem cell functionality are unknown. Here we demonstrate that type 2 innate immunity is required for regeneration of skeletal muscle after injury. Muscle damage results in rapid recruitment of eosinophils, which secrete IL-4 to activate the regenerative actions of muscle resident fibro/adipocyte progenitors (FAPs). In FAPs, IL-4/IL-13 signaling serves as a key switch to control their fate and functions. Activation of IL-4/IL-13 signaling promotes proliferation of FAPs to support myogenesis, while inhibiting their differentiation into adipocytes. Surprisingly, type 2 cytokine signaling is also required in FAPs, but not myeloid cells, for rapid clearance of necrotic debris, a process that is necessary for timely and complete regeneration of tissues.

The cocrystal structure of human poly(A)-binding protein (PABP) has been determined at 2.6 A resolution. PABP recognizes the 3' mRNA poly(A) tail and plays critical roles in eukaryotic translation initiation and mRNA stabilization/degradation. The minimal PABP used in this study consists of the N-terminal two RRM-type RNA-binding domains connected by a short linker (RRM1/2). These two RRMs form a continuous RNA-binding trough, lined by an antiparallel beta sheet backed by four alpha helices. The polyadenylate RNA adopts an extended conformation running the length of the molecular trough. Adenine recognition is primarily mediated by contacts with conserved residues found in the RNP motifs of the two RRMs. The convex dorsum of RRM1/2 displays a phylogenetically conserved hydrophobic/acidic portion, which may interact with translation initiation factors and regulatory proteins.

Supplemental Digital Content is available in the text.

Genotypic differences greatly influence susceptibility and resistance to disease. Understanding genotype-phenotype relationships requires that phenotypes be viewed as manifestations of network properties, rather than simply as the result of individual genomic variations1. Genome sequencing efforts have identified numerous germline mutations associated with cancer predisposition and large numbers of somatic genomic alterations2. However, it remains challenging to distinguish between background, or “passenger” and causal, or “driver” cancer mutations in these datasets. Human viruses intrinsically depend on their host cell during the course of infection and can elicit pathological phenotypes similar to those arising from mutations3. To test the hypothesis that genomic variations and tumour viruses may cause cancer via related mechanisms, we systematically examined host interactome and transcriptome network perturbations caused by DNA tumour virus proteins. The resulting integrated viral perturbation data reflects rewiring of the host cell networks, and highlights pathways that go awry in cancer, such as Notch signalling and apoptosis. We show that systematic analyses of host targets of viral proteins can identify cancer genes with a success rate on par with their identification through functional genomics and large-scale cataloguing of tumour mutations. Together, these complementary approaches result in increased specificity for cancer gene identification. Combining systems-level studies of pathogen-encoded gene products with genomic approaches will facilitate prioritization of cancer-causing driver genes so as to advance understanding of the genetic basis of human cancer.

Loss of kidney function underlies many renal diseases1. Mammals can partly repair their nephrons (the functional units of the kidney), but cannot form new ones2,3. By contrast, fish add nephrons throughout their lifespan and regenerate nephrons de novo after injury4,5, providing a model for understanding how mammalian renal regeneration may be therapeutically activated. Here we trace the source of new nephrons in the adult zebrafish to small cellular aggregates containing nephron progenitors. Transplantation of single aggregates comprising 10–30 cells is sufficient to engraft adults and generate multiple nephrons. Serial transplantation experiments to test self-renewal revealed that nephron progenitors are long-lived and possess significant replicative potential, consistent with stem-cell activity. Transplantation of mixed nephron progenitors tagged with either green or red fluorescent proteins yielded some mosaic nephrons, indicating that multiple nephron progenitors contribute to a single nephron. Consistent with this, live imaging of nephron formation in transparent larvae showed that nephrogenic aggregates form by the coalescence of multiple cells and then differentiate into nephrons. Taken together, these data demonstrate that the zebrafish kidney probably contains self-renewing nephron stem/progenitor cells. The identification of these cells paves the way to isolating or engineering the equivalent cells in mammals and developing novel renal regenerative therapies.

Exercise provides numerous salutary effects, but our understanding of how these occur is limited. To gain a clearer picture of exercise-induced metabolic responses, we have developed comprehensive plasma metabolite signatures by using mass spectrometry to measure over 200 metabolites before and after exercise. We identified plasma indicators of glycogenolysis (glucose-6-phosphate), tricarboxylic acid (TCA) cycle span 2 expansion (succinate, malate, and * To whom correspondence should be addressed Corresponding authors Robert E. Gerszten, MD Cardiology Division and Center for Immunology & Inflammatory Diseases Massachusetts General Hospital, Room 8307 149 13th Street Charlestown, MA 02129 rgerszten@partners.org Gregory D. Lewis, MD Cardiology Division Massachusetts General Hospital, GRB 800 55 Fruit Street, Boston, MA 02114 glewis@partners.org. Authors contributions: G.D.L conceived the study, designed the experiments, performed primary data analysis and wrote the manuscript. M.J.W. led the effort to recruit and phenotype marathon subjects, L.F. and M.M. recruited subjects, processed samples, and assisted with experimental design. Z.A. and G.C.R. designed and performed the gene expression profiling experiments, A.S., E.Y., X.S., A.A., S.A.C. and C.B.C. developed the metabolic profiling platform, performed mass spectrometry experiments, and analyzed the data, S.C., E.L.M, T.W., and R.S.V. designed experiments and analyzed data from the Framingham Heart Study cohort, R.D. and F.P.R. assisted with statistical analysis and constructed the metabolite interrelatedness dendrogram, E.P.R. contributed to mass spectrometry data analysis and helped to write the manuscript, D.M.S. and M.J.S. contributed to the cardiopulmonary exercise testing metabolic profiling experiment, M.S.S. helped to conceive and design the exercise treadmill testing studies and assisted in data interpretation and in writing the manuscript, R.E.G. conceived of the study, designed experiments, analyzed data, and wrote the manuscript. Competing interests:The authors declare that they have no competing interests. NIH Public Access Author ManuscriptSci Transl Med. Author manuscript; available in PMC 2010 December 27. NIH-PA Author ManuscriptNIH-PA Author Manuscript NIH-PA Author Manuscript fumarate), and lipolysis (glycerol), as well as modulators of insulin sensitivity (niacinamide) and fatty acid oxidation (pantothenic acid). Metabolites that were highly correlated with fitness parameters were found in subjects undergoing acute exercise testing, marathon running, and in 302 subjects from a longitudinal cohort study. Exercise-induced increases in glycerol were strongly related to fitness levels in normal individuals and were attenuated in subjects with myocardial ischemia. A combination of metabolites that increased in plasma in response to exercise (glycerol, niacinamide, glucose-6-phosphate, pantothenate, and succinate) upregulated the expression of nur77, a transcriptional regulator of glucose utilization and lipid metabolism genes in skeleta...