Robustness and reproducibility of simple and complex synthetic logic circuit designs using a DBTL loop.

Computational tools addressing various components of design-build-test-learn (DBTL) loops for the construction of synthetic genetic networks exist but do not generally cover the entire DBTL loop. This manuscript introduces an end-to-end sequence of tools that together form a DBTL loop called Design Assemble Round Trip (DART). DART provides rational selection and refinement of genetic parts to construct and test a circuit. Computational support for experimental process, metadata management, standardized data collection and reproducible data analysis is provided via the previously published Round Trip (RT) test-learn loop. The primary focus of this work is on the Design Assemble (DA) part of the tool chain, which improves on previous techniques by screening up to thousands of network topologies for robust performance using a novel robustness score derived from dynamical behavior based on circuit topology only. In addition, novel experimental support software is introduced for the assembly of genetic circuits. A complete design-through-analysis sequence is presented using several OR and NOR circuit designs, with and without structural redundancy, that are implemented in budding yeast. The execution of DART tested the predictions of the design tools, specifically with regard to robust and reproducible performance under different experimental conditions. The data analysis depended on a novel application of machine learning techniques to segment bimodal flow cytometry distributions. Evidence is presented that, in some cases, a more complex build may impart more robustness and reproducibility across experimental conditions. Graphical Abstract.

Neural network based integration of assays to assess pathogenic potential.

Limited data significantly hinders our capability of biothreat assessment of novel bacterial strains. Integration of data from additional sources that can provide context about the strain can address this challenge. Datasets from different sources, however, are generated with a specific objective and which makes integration challenging. Here, we developed a deep learning-based approach called the neural network embedding model (NNEM) that integrates data from conventional assays designed to classify species with new assays that interrogate hallmarks of pathogenicity for biothreat assessment. We used a dataset of metabolic characteristics from a de-identified set of known bacterial strains that the Special Bacteriology Reference Laboratory (SBRL) of the Centers for Disease Control and Prevention (CDC) has curated for use in species identification. The NNEM transformed results from SBRL assays into vectors to supplement unrelated pathogenicity assays from de-identified microbes. The enrichment resulted in a significant improvement in accuracy of 9% for biothreat. Importantly, the dataset used in our analysis is large, but noisy. Therefore, the performance of our system is expected to improve as additional types of pathogenicity assays are developed and deployed. The proposed NNEM strategy thus provides a generalizable framework for enrichment of datasets with previously collected assays indicative of species.

A toolkit for enhanced reproducibility of RNASeq analysis for synthetic biologists.

Sequencing technologies, in particular RNASeq, have become critical tools in the design, build, test and learn cycle of synthetic biology. They provide a better understanding of synthetic designs, and they help identify ways to improve and select designs. While these data are beneficial to design, their collection and analysis is a complex, multistep process that has implications on both discovery and reproducibility of experiments. Additionally, tool parameters, experimental metadata, normalization of data and standardization of file formats present challenges that are computationally intensive. This calls for high-throughput pipelines expressly designed to handle the combinatorial and longitudinal nature of synthetic biology. In this paper, we present a pipeline to maximize the analytical reproducibility of RNASeq for synthetic biologists. We also explore the impact of reproducibility on the validation of machine learning models. We present the design of a pipeline that combines traditional RNASeq data processing tools with structured metadata tracking to allow for the exploration of the combinatorial design in a high-throughput and reproducible manner. We then demonstrate utility via two different experiments: a control comparison experiment and a machine learning model experiment. The first experiment compares datasets collected from identical biological controls across multiple days for two different organisms. It shows that a reproducible experimental protocol for one organism does not guarantee reproducibility in another. The second experiment quantifies the differences in experimental runs from multiple perspectives. It shows that the lack of reproducibility from these different perspectives can place an upper bound on the validation of machine learning models trained on RNASeq data. Graphical Abstract.

Phenotype-Based Threat Assessment.

Bacterial pathogen identification, which is critical for human health, has historically relied on culturing organisms from clinical specimens. More recently, the application of machine learning (ML) to whole-genome sequences (WGSs) has facilitated pathogen identification. However, relying solely on genetic information to identify emerging or new pathogens is fundamentally constrained, especially if novel virulence factors exist. In addition, even WGSs with ML pipelines are unable to discern phenotypes associated with cryptic genetic loci linked to virulence. Here, we set out to determine if ML using phenotypic hallmarks of pathogenesis could assess potential pathogenic threat without using any sequence-based analysis. This approach successfully classified potential pathogenetic threat associated with previously machine-observed and unobserved bacteria with 99% and 85% accuracy, respectively. This work establishes a phenotype-based pipeline for potential pathogenic threat assessment, which we term PathEngine, and offers strategies for the identification of bacterial pathogens.

Prediction of whole-cell transcriptional response with machine learning.

MOTIVATION: Applications in synthetic and systems biology can benefit from measuring whole-cell response to biochemical perturbations. Execution of experiments to cover all possible combinations of perturbations is infeasible. In this paper, we present the host response model (HRM), a machine learning approach that maps response of single perturbations to transcriptional response of the combination of perturbations. RESULTS: The HRM combines high-throughput sequencing with machine learning to infer links between experimental context, prior knowledge of cell regulatory networks, and RNASeq data to predict a gene's dysregulation. We find that the HRM can predict the directionality of dysregulation to a combination of inducers with an accuracy of >90% using data from single inducers. We further find that the use of prior, known cell regulatory networks doubles the predictive performance of the HRM (an R2 from 0.3 to 0.65). The model was validated in two organisms, Escherichia coli and Bacillus subtilis, using new experiments conducted after training. Finally, while the HRM is trained with gene expression data, the direct prediction of differential expression makes it possible to also conduct enrichment analyses using its predictions. We show that the HRM can accurately classify >95% of the pathway regulations. The HRM reduces the number of RNASeq experiments needed as responses can be tested in silico prior to the experiment. AVAILABILITY AND IMPLEMENTATION: The HRM software and tutorial are available at https://github.com/sd2e/CDM and the configurable differential expression analysis tools and tutorials are available at https://github.com/SD2E/omics_tools. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

RNA Solutions: Synthesizing Information to Support Transcriptomics (RNASSIST).

MOTIVATION: Transcriptomics is a common approach to identify changes in gene expression induced by a disease state. Standard transcriptomic analyses consider differentially expressed genes (DEGs) as indicative of disease states so only a few genes would be treated as signals when the effect size is small, such as in brain tissue. For tissue with small effect sizes, if the DEGs do not belong to a pathway known to be involved in the disease, there would be little left in the transcriptome for researchers to follow up with. RESULTS: We developed RNA Solutions: Synthesizing Information to Support Transcriptomics (RNASSIST), a new approach to identify hidden signals in transcriptomic data by linking differential expression and co-expression networks using machine learning. We applied our approach to RNA-seq data of post-mortem brains that compared the Alcohol Use Disorder (AUD) group with the control group. Many of the candidate genes are not differentially expressed so would likely be ignored by standard transcriptomic analysis pipelines. Through multiple validation strategies, we concluded that these RNASSIST-identified genes likely play a significant role in AUD. AVAILABILITY AND IMPLEMENTATION: The RNASSIST algorithm is available at https://github.com/netrias/rnassist and both the software and the data used in RNASSIST are available at https://figshare.com/articles/software/RNAssist_Software_and_Data/16617250. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

A single institutional experience with suction thrombectomy in acute mesenteric ischemia.

Acute mesenteric ischemia (AMI) is typically treated by open surgery or hybrid techniques. Catheter-based aspiration thrombectomy represents another minimally invasive alternative with a potential additional safety benefit of minimizing the bleeding risk associated with thrombolytics. In this institutional case series, we present five clinical cases of aspiration thrombectomy for high-risk AMI using the Penumbra aspiration system. All patients underwent technically successful endovascular thrombectomy as demonstrated by intraoperative angiography results. However, bowel necrosis and sepsis adversely affected postoperative outcomes. Lack of intraoperative bowel assessment is a limitation of endovascular methods, highlighting the importance of patient selection.

Situated Precision Healthcare in the Smart Medical Home: Bringing NASA's Research Strategy down to Earth.

This special issue is ambitious in that it calls for strategic transformation in research on Alzheimer's Disease (AD) and related dementias, including innovation in both research design and value delivery, through lifestyle interventions that implicitly relate to a much broader range of comorbidities and diseases of aging. One response to this challenge is to venture beyond the boundaries of research that supports the healthcare industry. Toward this end, we introduce opportunities for research translation and knowledge transfer from NASA to the healthcare industry. Our intent is to show how NASA's approach to research can guide innovation for a smart medical home, most notably for AD and other diseases of aging. The article is organized in four major sections: (a) aggregating fragmented research communities; (b) lifestyle interventions in the medical home; (c) multiscale computational modeling and analysis; and (d) lifespan approach to precision brain health. We provide novel motivations and transformative paths to a diversity of specific lines of research, across communities, that would be difficult to discover in common methods of networking within research communities and even through sophisticated bibliographic methods. We thus reveal how knowledge transfer between the public and private sector can stimulate development of broader scientific communities and achieve a more coherent strategic approach to integration and development of a diversity of capabilities including but not limited to technology.

Validation of a preoperative prediction model for mortality within 1 year after endovascular aortic aneurysm repair of intact aneurysms.

OBJECTIVE: Most would agree that at least 1-year survival is necessary after intact abdominal aortic aneurysm (AAA) repair to appropriately justify the cost and risk of the procedure. No validated clinical decision instruments exist to predict survival after endovascular aneurysm repair (EVAR) beyond the perioperative period. The purpose of this analysis was to create a preoperative prediction model for 1-year mortality after EVAR for intact AAA in the Society for Vascular Surgery Vascular Quality Initiative. METHODS: All intact EVARs in the Society for Vascular Surgery Vascular Quality Initiative from 2011 to 2015 were randomly divided into training (n = 17,836) and validation (n = 2500) data sets, and 31 preoperative candidate predictors were identified. A logistic regression model for 1-year mortality was created, and bootstrapped stepwise variable elimination was used to reduce this model to a best subset of predictors. Penalized maximum likelihood estimation was used to correct for potential overfitting. The final model was internally validated by bootstrapping the area under the curve (AUC) and the calibration slope and intercept, and its performance when applied to the training and validation data sets was compared. RESULTS: After elective and nonelective (symptomatic, intact) EVAR, 1-year mortality was 5.5% (n = 900/16,411) and 11.4% (n = 162/1425), respectively. The mean probability of 1-year mortality was 6.0% (n = 1062) in the training set and 5.7% (n = 143) in the validation cohort (P = .12). Significant preoperative predictors of 1-year mortality included chronic obstructive pulmonary disease, age, preoperative renal insufficiency (creatinine concentration ≥1.8 mg/dL or on hemodialysis), ejection fraction <50%, transfer status, body mass index <24 kg/m2, preoperative beta-blocker exposure, larger AAA diameter, and lower admission hemoglobin level. Preoperative statin use was found to be protective. The bias-corrected AUC was 0.759 (Hosmer-Lemeshow goodness-of-fit P value of 0.36; calibration intercept, -0.003; slope, 0.999). When applied to the validation data set, the model had AUC of 0.724 (95% confidence interval, 0.676-0.768; calibration intercept, 0.0009; slope, 0.970), which was in excellent agreement with the original data set bias-corrected AUC. Notably, ∼27.5% (n = 4902) had four or more risk factors with a predicted 1-year post-EVAR mortality risk of 10% to 22% despite that 33.2% of these patients had AAA diameters below recommended treatment guideline minimum thresholds. CONCLUSIONS: This validated preoperative prediction model for 1-year mortality identifies patients less likely to benefit from EVAR. Appropriateness of intact AAA EVAR care delivery can be improved by use of this clinical decision aid to determine which high-risk patients have lower probability of mortality within the first postoperative year relative to their predicted annualized rupture risk.

Outcomes of patients with type I endoleak at completion of endovascular abdominal aneurysm repair.

BACKGROUND: Type I endoleak (TIE) during endovascular aneurysm repair (EVAR) is usually identified and treated intraoperatively. We evaluated the outcomes of patients who, despite possible treatment, had TIE at completion of EVAR. METHODS: We examined consecutive EVAR for nonruptured abdominal aortic aneurysm (AAA) within the Vascular Study Group of New England database (2003-2012) and compared the outcomes of patients who had TIE at completion with those who did not. Outcomes included perioperative death, cardiac complication, reoperation, and 1-year mortality. Multivariable logistic regression was used to determine factors associated with perioperative mortality, as well as factors associated with TIE. Anatomic factors associated with TIE were not evaluated because of the limitations of the Vascular Study Group of New England database. RESULTS: Among the 2402 EVARs for nonruptured AAA in the Vascular Study Group of New England sample, 93% (n = 2235) were performed electively and 7% had (n = 167) symptomatic AAA. Eighty patients (3.3%) had TIE at completion of surgery. Patients with TIE were older (77.9 vs 73.9 years; P < .001), had higher female preponderance (34% vs 20%; P = .004), larger endograft main body diameter (28.8 vs 27.2 mm; P < .001), and more unplanned graft extension (32% vs 10%; P < .001) than those without TIE. At 1-year follow-up, 90% of patients who had TIE at the completion of their EVAR had resolution of TIE without further need for endovascular intervention or open conversion type I endoleak at the completion of surgery was associated with increased in-hospital mortality (5% vs 0.6%; P = .002) and cardiac dysrhythmia (8.8% vs 3.2%; P = .02). In multivariable analysis, TIE was independently associated with increased odds of in-hospital mortality (odds ratio [OR], 4.4; 95% confidence interval [CI], 1.2-16.4; P = .03). Multivariable analysis revealed the following factors to be independently predictive of TIE: female gender (OR, 2.2, 95% CI, 1.3-3.7; P = .002), patients older than 70 years of age (OR, 2.0; 95% CI, 1.1-3.8; P = .02), those with main body graft diameter >30 mm (OR, 2.6; 95% CI, 1.6-4.3; P < .001), and those undergoing unplanned graft extension (OR, 4.6; 95% CI, 2.7-7.7; P < .001). CONCLUSIONS: TIE occurred in 3% of patients at completion of EVAR with more than 90% resolved spontaneously at 1-year follow-up. It is associated with increased risk of in-hospital mortality and cardiac complication. Additional investigation is needed to further define anatomic factors associated with TIE and to improve perioperative outcomes of these at-risk patients.

A randomized comparison of a 3-week and 6-week vascular surgery simulation course on junior surgical residents' performance of an end-to-side anastomosis.

OBJECTIVE: We assessed the effect of an open vascular simulation course on the surgical skill of junior surgical residents in performing a vascular end-to-side anastomosis and determined the course length required for effectiveness. We hypothesized that a 6-week course would significantly increase the surgical skill of junior residents in performing an end-to-side anastomosis, while a 3-week course would not. METHODS: We randomized 37 junior residents (postgraduate year 1 to 3) to a course consisting of three (short course, n = 18) or six (long course, n = 19) consecutive weekly 1-hour teaching sessions. Content focused on instrument recognition and performance of an end-to-side vascular anastomosis using a simulation model. A standardized 50-point vascular skills assessment (SVSA) measured knowledge and technical proficiency. Senior residents (postgraduate year 4 to 5) were tested at baseline. Junior residents were tested at baseline and at 1 and 16 weeks after course completion, and their scores were compared with baseline and senior resident scores. Residents and faculty completed a standardized anonymous evaluation of the course. RESULTS: Baseline scores between short-course and long-course participants were not different. At baseline, junior residents had significantly lower SVSA scores than senior residents (36±7 vs 41.4±2.5; P=.002). One week after course completion, SVSA scores for short-course (43.5±2.9 vs 34.2±7.5; P=.008) and long-course (43.9±5.6 vs 38.3±5.9; P=.006) participants were significantly improved from baseline. SVSA scores decreased slightly at 16 weeks but remained above baseline in short-course (39±6.2 vs 34.2±7.5; P=.03) and long-course (40±4.5 vs 38.3±5.9; P=.08) participants. Long vs short course length did not affect improvement in SVSA scores at 1 or 16 weeks. In short-course and long-course participants, SVSA scores at 1 and 16 weeks were not significantly different from senior resident scores. Course ratings were high, and 95% of residents indicated the course "made them a better surgeon." Residents and faculty felt the educational benefit of the course merited the investment of resources. CONCLUSIONS: An open vascular simulation course consisting of three weekly 1-hour sessions increased the surgical skill of junior residents in performing a vascular end-to-side anastomosis to that of senior residents on a standardized assessment. A 6-week course provided no additional benefit. This study supports the use of an open vascular simulation course to teach vascular surgical skills to junior residents. A course consisting of three 1-hour sessions is an effective and efficient component of a simulation program for junior surgical residents in a busy surgical center.