Using public clinical trial reports to probe non-experimental causal inference methods.

BACKGROUND: Non-experimental studies (also known as observational studies) are valuable for estimating the effects of various medical interventions, but are notoriously difficult to evaluate because the methods used in non-experimental studies require untestable assumptions. This lack of intrinsic verifiability makes it difficult both to compare different non-experimental study methods and to trust the results of any particular non-experimental study. METHODS: We introduce TrialProbe, a data resource and statistical framework for the evaluation of non-experimental methods. We first collect a dataset of pseudo "ground truths" about the relative effects of drugs by using empirical Bayesian techniques to analyze adverse events recorded in public clinical trial reports. We then develop a framework for evaluating non-experimental methods against that ground truth by measuring concordance between the non-experimental effect estimates and the estimates derived from clinical trials. As a demonstration of our approach, we also perform an example methods evaluation between propensity score matching, inverse propensity score weighting, and an unadjusted approach on a large national insurance claims dataset. RESULTS: From the 33,701 clinical trial records in our version of the ClinicalTrials.gov dataset, we are able to extract 12,967 unique drug/drug adverse event comparisons to form a ground truth set. During our corresponding methods evaluation, we are able to use that reference set to demonstrate that both propensity score matching and inverse propensity score weighting can produce estimates that have high concordance with clinical trial results and substantially outperform an unadjusted baseline. CONCLUSIONS: We find that TrialProbe is an effective approach for probing non-experimental study methods, being able to generate large ground truth sets that are able to distinguish how well non-experimental methods perform in real world observational data.

Data-driven hypothesis weighting increases detection power in genome-scale multiple testing.

Hypothesis weighting improves the power of large-scale multiple testing. We describe independent hypothesis weighting (IHW), a method that assigns weights using covariates independent of the P-values under the null hypothesis but informative of each test's power or prior probability of the null hypothesis (http://www.bioconductor.org/packages/IHW). IHW increases power while controlling the false discovery rate and is a practical approach to discovering associations in genomics, high-throughput biology and other large data sets.

Mapping lung cancer epithelial-mesenchymal transition states and trajectories with single-cell resolution.

Elucidating the spectrum of epithelial-mesenchymal transition (EMT) and mesenchymal-epithelial transition (MET) states in clinical samples promises insights on cancer progression and drug resistance. Using mass cytometry time-course analysis, we resolve lung cancer EMT states through TGFß-treatment and identify, through TGFß-withdrawal, a distinct MET state. We demonstrate significant differences between EMT and MET trajectories using a computational tool (TRACER) for reconstructing trajectories between cell states. In addition, we construct a lung cancer reference map of EMT and MET states referred to as the EMT-MET PHENOtypic STAte MaP (PHENOSTAMP). Using a neural net algorithm, we project clinical samples onto the EMT-MET PHENOSTAMP to characterize their phenotypic profile with single-cell resolution in terms of our in vitro EMT-MET analysis. In summary, we provide a framework to phenotypically characterize clinical samples in the context of in vitro EMT-MET findings which could help assess clinical relevance of EMT in cancer in future studies.

Creating functional engineered variants of the single-module non-ribosomal peptide synthetase IndC by T domain exchange.

Non-ribosomal peptide synthetases (NRPSs) are enzymes that catalyze ribosome-independent production of small peptides, most of which are bioactive. NRPSs act as peptide assembly lines where individual, often interconnected modules each incorporate a specific amino acid into the nascent chain. The modules themselves consist of several domains that function in the activation, modification and condensation of the substrate. NRPSs are evidently modular, yet experimental proof of the ability to engineer desired permutations of domains and modules is still sought. Here, we use a synthetic-biology approach to create a small library of engineered NRPSs, in which the domain responsible for carrying the activated amino acid (T domain) is exchanged with natural or synthetic T domains. As a model system, we employ the single-module NRPS IndC from Photorhabdus luminescens that produces the blue pigment indigoidine. As chassis we use Escherichia coli. We demonstrate that heterologous T domain exchange is possible, even for T domains derived from different organisms. Interestingly, substitution of the native T domain with a synthetic one enhanced indigoidine production. Moreover, we show that selection of appropriate inter-domain linker regions is critical for functionality. Taken together, our results extend the engineering avenues for NRPSs, as they point out the possibility of combining domain sequences coming from different pathways, organisms or from conservation criteria. Moreover, our data suggest that NRPSs can be rationally engineered to control the level of production of the corresponding peptides. This could have important implications for industrial and medical applications.