Selective protein kinase inhibitors were developed on the basis of the unexpected binding mode of 2,6,9-trisubstituted purines to the adenosine triphosphate-binding site of the human cyclin-dependent kinase 2 (CDK2). By iterating chemical library synthesis and biological screening, potent inhibitors of the human CDK2-cyclin A kinase complex and of Saccharomyces cerevisiae Cdc28p were identified. The structural basis for the binding affinity and selectivity was determined by analysis of a three-dimensional crystal structure of a CDK2-inhibitor complex. The cellular effects of these compounds were characterized in mammalian cells and yeast. In the latter case the effects were characterized on a genome-wide scale by monitoring changes in messenger RNA levels in treated cells with high-density oligonucleotide probe arrays. Purine libraries could provide useful tools for analyzing a variety of signaling and regulatory pathways and may lead to the development of new therapeutics.

The detection of somatic mutations from cancer genome sequences is key to understanding the genetic basis of disease progression, patient survival and response to therapy. Benchmarking is needed for tool assessment and improvement but is complicated by a lack of gold standards, by extensive resource requirements and by difficulties in sharing personal genomic information. To resolve these issues, we launched the ICGC-TCGA DREAM Somatic Mutation Calling Challenge, a crowdsourced benchmark of somatic mutation detection algorithms. Here we report the BAMSurgeon tool for simulating cancer genomes and the results of 248 analyses of three in silico tumors created with it. Different algorithms exhibit characteristic error profiles, and, intriguingly, false positives show a trinucleotide profile very similar to one found in human tumors. Although the three simulated tumors differ in sequence contamination (deviation from normal cell sequence) and in subclonality, an ensemble of pipelines outperforms the best individual pipeline in all cases. BAMSurgeon is available at https://github.com/adamewing/bamsurgeon/.

The effectiveness of most cancer targeted therapies is short-lived. Tumors often develop resistance that might be overcome with drug combinations. However, the number of possible combinations is vast, necessitating data-driven approaches to find optimal patient-specific treatments. Here we report AstraZeneca’s large drug combination dataset, consisting of 11,576 experiments from 910 combinations across 85 molecularly characterized cancer cell lines, and results of a DREAM Challenge to evaluate computational strategies for predicting synergistic drug pairs and biomarkers. 160 teams participated to provide a comprehensive methodological development and benchmarking. Winning methods incorporate prior knowledge of drug-target interactions. Synergy is predicted with an accuracy matching biological replicates for >60% of combinations. However, 20% of drug combinations are poorly predicted by all methods. Genomic rationale for synergy predictions are identified, including ADAM17 inhibitor antagonism when combined with PIK3CB/D inhibition contrasting to synergy when combined with other PI3K-pathway inhibitors in PIK3CA mutant cells.

2,6,9-Trisubstituted purines have various and potent biological activities, despite high concentrations of competing endogenous purine ligands in living cells. Purine libraries constitute a versatile source of small molecules that affect distinct biochemical pathways mediating different cellular functions.

IMPORTANCE Mammography screening currently relies on subjective human interpretation. Artificial intelligence (AI) advances could be used to increase mammography screening accuracy by reducing missed cancers and false positives. OBJECTIVE To evaluate whether AI can overcome human mammography interpretation limitations with a rigorous, unbiased evaluation of machine learning algorithms. DESIGN, SETTING, AND PARTICIPANTS In this diagnostic accuracy study conducted between September 2016 and November 2017, an international, crowdsourced challenge was hosted to foster AI algorithm development focused on interpreting screening mammography. More than 1100 participants comprising 126 teams from 44 countries participated. Analysis began November 18, 2016. MAIN OUTCOMES AND MEASUREMENTS Algorithms used images alone (challenge 1) or combined images, previous examinations (if available), and clinical and demographic risk factor data (challenge 2) and output a score that translated to cancer yes/no within 12 months. Algorithm accuracy for breast cancer detection was evaluated using area under the curve and algorithm specificity compared with radiologists' specificity with radiologists' sensitivity set at 85.9% (United States) and 83.9% (Sweden). An ensemble method aggregating top-performing AI algorithms and radiologists' recall assessment was developed and evaluated. RESULTS Overall, 144 231 screening mammograms from 85 580 US women (952 cancer positive Յ12 months from screening) were used for algorithm training and validation. A second independent validation cohort included 166 578 examinations from 68 008 Swedish women (780 cancer positive). The top-performing algorithm achieved an area under the curve of 0.858 (United States) and 0.903 (Sweden) and 66.2% (United States) and 81.2% (Sweden) specificity at the radiologists' sensitivity, lower than community-practice radiologists' specificity of 90.5% (United States) and 98.5% (Sweden). Combining top-performing algorithms and US radiologist assessments resulted in a higher area under the curve of 0.942 and achieved a significantly improved specificity (92.0%) at the same sensitivity. CONCLUSIONS AND RELEVANCE While no single AI algorithm outperformed radiologists, an ensemble of AI algorithms combined with radiologist assessment in a single-reader screening environment improved overall accuracy. This study underscores the potential of using machine (continued)

Inferring molecular networks is a central challenge in computational biology. However, it has remained unclear whether causal, rather than merely correlational, relationships can be effectively inferred in complex biological settings. Here we describe the HPN-DREAM network inference challenge that focused on learning causal influences in signaling networks. We used phosphoprotein data from cancer cell lines as well as in silico data from a nonlinear dynamical model. Using the phosphoprotein data, we scored more than 2,000 networks submitted by challenge participants. The networks spanned 32 biological contexts and were scored in terms of causal validity with respect to unseen interventional data. A number of approaches were effective and incorporating known biology was generally advantageous. Additional sub-challenges considered time-course prediction and visualization. Our results constitute the most comprehensive assessment of causal network inference in a mammalian setting carried out to date and suggest that learning causal relationships may be feasible in complex settings such as disease states. Furthermore, our scoring approach provides a practical way to empirically assess the causal validity of inferred molecular networks.

The ribosomal incorporation of nonnative amino acids into polypeptides in living cells provides the opportunity to endow therapeutic proteins with unique pharmacological properties. We report here the first clinical study of a biosynthetic protein produced using an expanded genetic code. Incorporation of p-acetylphenylalanine (pAcF) at distinct locations in human growth hormone (hGH) allowed site-specific conjugation with polyethylene glycol (PEG) to produce homogeneous hGH variants. A mono-PEGylated mutant hGH modified at residue 35 demonstrated favorable pharmacodynamic properties in GH-deficient rats. Clinical studies in GH-deficient adults demonstrated efficacy and safety comparable to native human growth hormone therapy but with increased potency and reduced injection frequency. This example illustrates the utility of nonnative amino acids to optimize protein therapeutics in an analogous fashion to the use of medicinal chemistry to optimize conventional natural products, low molecular weight drugs, and peptides.protein engineering | endocrinology | bio-better

Fibroblast growth factor 21 (FGF21) mitigates many of the pathogenic features of type 2 diabetes, despite a short circulating half-life. PEGylation is a proven approach to prolonging the duration of action while enhancing biophysical solubility and stability. However, in the absence of a specific protein PEGylation site, chemical conjugation is inherently heterogeneous and commonly leads to dramatic loss in bioactivity. This work illustrates a novel means of specific PEGylation, producing FGF21 analogs with high specific activity and salutary biological activities. Using homology modeling and structure-based design, specific sites were chosen in human FGF21 for site-specific PEGylation to ensure that receptor binding regions were preserved. The in vitro activity of the PEGylated FGF21 ana-logs corresponded with the site of PEG placement within the binding model. Site-specific PEGylated analogs demonstrated dramatically increased circulating half-life and enhanced efficacy in db/db mice. Twice-weekly dosing of an optimal FGF21 analog reduced blood glucose, plasma lipids, liver triglycerides, and plasma glucagon and enhanced pancreatic insulin content, islet number, and glucose-dependent insulin secretion. Restoration of insulin sensitivity was demonstrated by the enhanced ability of insulin to induce Akt/protein kinase B phosphorylation in liver, muscle, and adipose tissues. PEGylation of human FGF21 at a specific and preferred site confers superior metabolic pharmacology.