Biologically informed NeuralODEs for genome-wide regulatory dynamics.

BACKGROUND: Gene regulatory network (GRN) models that are formulated as ordinary differential equations (ODEs) can accurately explain temporal gene expression patterns and promise to yield new insights into important cellular processes, disease progression, and intervention design. Learning such gene regulatory ODEs is challenging, since we want to predict the evolution of gene expression in a way that accurately encodes the underlying GRN governing the dynamics and the nonlinear functional relationships between genes. Most widely used ODE estimation methods either impose too many parametric restrictions or are not guided by meaningful biological insights, both of which impede either scalability, explainability, or both. RESULTS: We developed PHOENIX, a modeling framework based on neural ordinary differential equations (NeuralODEs) and Hill-Langmuir kinetics, that overcomes limitations of other methods by flexibly incorporating prior domain knowledge and biological constraints to promote sparse, biologically interpretable representations of GRN ODEs. We tested the accuracy of PHOENIX in a series of in silico experiments, benchmarking it against several currently used tools. We demonstrated PHOENIX's flexibility by modeling regulation of oscillating expression profiles obtained from synchronized yeast cells. We also assessed the scalability of PHOENIX by modeling genome-scale GRNs for breast cancer samples ordered in pseudotime and for B cells treated with Rituximab. CONCLUSIONS: PHOENIX uses a combination of user-defined prior knowledge and functional forms from systems biology to encode biological "first principles" as soft constraints on the GRN allowing us to predict subsequent gene expression patterns in a biologically explainable manner.

Biologically informed NeuralODEs for genome-wide regulatory dynamics.

Modeling dynamics of gene regulatory networks using ordinary differential equations (ODEs) allow a deeper understanding of disease progression and response to therapy, thus aiding in intervention optimization. Although there exist methods to infer regulatory ODEs, these are generally limited to small networks, rely on dimensional reduction, or impose non-biological parametric restrictions - all impeding scalability and explainability. PHOENIX is a neural ODE framework incorporating prior domain knowledge as soft constraints to infer sparse, biologically interpretable dynamics. Extensive experiments - on simulated and real data - demonstrate PHOENIX's unique ability to learn key regulatory dynamics while scaling to the whole genome.

An estimation of the endoscopist's musculoskeletal injury risk for right and left lateral decubitus positions during colonoscopy: a field-based ergonomic study.

BACKGROUND: Colonoscopy exposes endoscopists to awkward postures and prolonged forces, which increases their risk of musculoskeletal injury. Patient positioning has a significant impact on the ergonomics of colonoscopy. Recent trials have found the right lateral decubitus position is associated with quicker insertion, higher adenoma detection rates, and greater patient comfort compared to the left lateral decubitus position. However, this patient position is perceived as more strenuous by endoscopists. METHODS: Nineteen endoscopists were observed performing colonoscopies during a series of four-hour endoscopy clinics. Durations of each patient position (right lateral decubitus, left lateral decubitus, prone, and supine) were recorded for all observed procedures (n = 64). Endoscopist injury risk was estimated by a trained researcher for the first and last colonoscopies of the shifts (n = 34) using Rapid Upper Limb Assessment (RULA), an observational ergonomic tool that estimates risk of musculoskeletal injury by scoring postures of the upper body and factors such as muscle use, force, and load. The total RULA scores were compared with a Wilcoxon Signed-Rank test for patient position (right and left lateral decubitus) and time (first and last procedures) with significance taken at p < 0.05. Endoscopist preferences were also surveyed. RESULTS: The right lateral decubitus position was associated with significantly higher RULA scores than the left lateral decubitus position (median 5 vs. 3, p < 0.001). RULA scores were not significantly different between the first and last procedures of the shifts (median 5 vs. 5, p = 0.816). 89% of endoscopists preferred the left lateral decubitus position, primarily due to superior ergonomics and comfort. CONCLUSION: RULA scores indicate an increased risk of musculoskeletal injury in both patient positions, with greater risk in the right lateral decubitus position.

Biologically informed NeuralODEs for genome-wide regulatory dynamics.

Models that are formulated as ordinary differential equations (ODEs) can accurately explain temporal gene expression patterns and promise to yield new insights into important cellular processes, disease progression, and intervention design. Learning such ODEs is challenging, since we want to predict the evolution of gene expression in a way that accurately encodes the causal gene-regulatory network (GRN) governing the dynamics and the nonlinear functional relationships between genes. Most widely used ODE estimation methods either impose too many parametric restrictions or are not guided by meaningful biological insights, both of which impedes scalability and/or explainability. To overcome these limitations, we developed PHOENIX, a modeling framework based on neural ordinary differential equations (NeuralODEs) and Hill-Langmuir kinetics, that can flexibly incorporate prior domain knowledge and biological constraints to promote sparse, biologically interpretable representations of ODEs. We test accuracy of PHOENIX in a series of in silico experiments benchmarking it against several currently used tools for ODE estimation. We also demonstrate PHOENIX's flexibility by studying oscillating expression data from synchronized yeast cells and assess its scalability by modelling genome-scale breast cancer expression for samples ordered in pseudotime. Finally, we show how the combination of user-defined prior knowledge and functional forms from systems biology allows PHOENIX to encode key properties of the underlying GRN, and subsequently predict expression patterns in a biologically explainable way.

Ursodeoxycholic acid for prevention of gallstone disease after laparoscopic sleeve gastrectomy: an Atlantic Canada perspective.

BACKGROUND: Prophylactic ursodeoxycholic acid (UDCA) may be beneficial in reducing gallstone disease after bariatric surgery. The American Society for Metabolic and Bariatric Surgery (ASMBS) 2019 guidelines recommend a 6-month course of UDCA for patients undergoing laparoscopic sleeve gastrectomy (LSG). This has not been adopted broadly. This study intends to assess the effect of routine UDCA administration following LSG on symptomatic gallstone disease. METHODS: We performed a retrospective chart review of patients who underwent LSG, between 2009 and 2019, at two tertiary care centers in Atlantic Canada. At one center, UDCA 250 mg oral twice daily was routinely prescribed following LSG for 6 months to patients with an intact gallbladder. At the other center, UDCA was not prescribed. Primary and secondary outcomes were cholecystectomy and endoscopic retrograde cholangiopancreatography (ERCP) rates. Compliance with and side effects of UDCA therapy were analyzed. RESULTS: A total of 751 patients were included in the study. Patients who had prior cholecystectomy or were lost to follow up were excluded. After exclusion criteria were applied, 461 patients were included for analysis: 303 in the UDCA group and 158 in the group who did not receive UDCA. Cholecystectomy rate was not significantly associated with UDCA administration, however there was a trend towards less cholecystectomy in patients who received UDCA (8.3% vs. 13.9%, p = 0.056). ERCP rate was significantly lower in patients who received UDCA (0.3% vs 2.5%, p = 0.031). Rate of gallstone disease requiring intervention, either cholecystectomy or ERCP, was significantly decreased in patients who received UDCA (8.9% vs 15.8%, p = 0.022). The most common barriers to compliance with UDCA were cost (45.4%) and nausea (18.1%). CONCLUSION: This is the first study to demonstrate lower rates of ERCP in patients receiving routine UDCA following LSG. Our findings support the ASMBS 2019 guidelines for administering UDCA after LSG for preventing gallstone disease.

The absence of the queuosine tRNA modification leads to pleiotropic phenotypes revealing perturbations of metal and oxidative stress homeostasis in Escherichia coli K12.

Queuosine (Q) is a conserved hypermodification of the wobble base of tRNA containing GUN anticodons but the physiological consequences of Q deficiency are poorly understood in bacteria. This work combines transcriptomic, proteomic and physiological studies to characterize a Q-deficient Escherichia coli K12 MG1655 mutant. The absence of Q led to an increased resistance to nickel and cobalt, and to an increased sensitivity to cadmium, compared to the wild-type (WT) strain. Transcriptomic analysis of the WT and Q-deficient strains, grown in the presence and absence of nickel, revealed that the nickel transporter genes (nikABCDE) are downregulated in the Q- mutant, even when nickel is not added. This mutant is therefore primed to resist to high nickel levels. Downstream analysis of the transcriptomic data suggested that the absence of Q triggers an atypical oxidative stress response, confirmed by the detection of slightly elevated reactive oxygen species (ROS) levels in the mutant, increased sensitivity to hydrogen peroxide and paraquat, and a subtle growth phenotype in a strain prone to accumulation of ROS.

Morbid Obesity and Severe Knee Osteoarthritis: Which Should Be Treated First?

BACKGROUND: There are limited prospective data, and conflicting retrospective data, providing guidance on how to optimally manage patients with morbid obesity and severe knee osteoarthritis. This study sought to review the effect of bariatric surgery on knee pain and knee surgery 30-day outcomes, as well as assess whether the sequence of bariatric and knee surgery has any effect on 30-day complications. METHODS: A retrospective chart review of all patients undergoing laparoscopic sleeve gastrectomy (LSG) from July 2006 to July 2016 at a university hospital was performed. Patients with knee pain or knee surgery (pre- or post-LSG) were identified using bariatric and orthopedic clinic notes. Those who had improvement in knee pain following LSG resulting in removal from orthopedic surgery waitlist were identified. We also assessed surgical outcomes in patients undergoing knee arthroscopy or total knee arthroplasty (TKA) followed by LSG compared to patients undergoing LSG followed by knee arthroscopy or TKA. RESULTS: During our study timeframe, 355 patients underwent LSG. Knee pain was documented in 150 (42.2%) patients, and orthopedic surgery consultation was completed for 57 (38.0%) patients with knee pain. Orthopedic intervention was performed prior to LSG for 24 patients and after LSG for 27 patients. Procedures were a combination of arthroscopy (18) and TKA (33). Six patients were removed from the waitlist for TKA following LSG due to resolution of symptoms. Order of interventions did not affect 30-day complications for patients undergoing LSG and arthroscopy (16% arthroscopy first, 0% LSG first, p = 0.43). A higher rate of LSG complications was noted in patients who underwent TKA prior to LSG (25% vs 0%, p = 0.04). There were no differences in TKA complications (8.3% TKA first, 4.8% LSG first, p = 1.00). CONCLUSION: In a small number of patients with morbid obesity and severe knee osteoarthritis, orthopedic intervention can be delayed and potentially avoided by undergoing LSG. In our study, 6/57 (10.5%) of patients with orthopedic consultation prior to LSG saw resolution of symptoms of knee pain. Referral to bariatric surgery should be considered for patients with morbid obesity and severe knee osteoarthritis.

Adverse Outcomes in Obese Cardiac Surgery Patients Correlates With Altered Branched-Chain Amino Acid Catabolism in Adipose Tissue and Heart.

Background: Predicting relapses of post-operative complications in obese patients who undergo cardiac surgery is significantly complicated by persistent metabolic maladaptation associated with obesity. Despite studies supporting the linkages of increased systemic branched-chain amino acids (BCAAs) driving the pathogenesis of obesity, metabolome wide studies have either supported or challenged association of circulating BCAAs with cardiovascular diseases (CVDs). Objective: We interrogated whether BCAA catabolic changes precipitated by obesity in the heart and adipose tissue can be reliable prognosticators of adverse outcomes following cardiac surgery. Our study specifically clarified the correlation between BCAA catabolizing enzymes, cellular BCAAs and branched-chain keto acids (BCKAs) with the severity of cardiometabolic outcomes in obese patients pre and post cardiac surgery. Methods: Male and female patients of ages between 44 and 75 were stratified across different body mass index (BMI) (non-obese = 17, pre-obese = 19, obese class I = 14, class II = 17, class III = 12) and blood, atrial appendage (AA), and subcutaneous adipose tissue (SAT) collected during cardiac surgery. Plasma and intracellular BCAAs and BC ketoacids (BCKAs), tissue mRNA and protein expression and activity of BCAA catabolizing enzymes were assessed and correlated with clinical parameters. Results: Intramyocellular, but not systemic, BCAAs increased with BMI in cardiac surgery patients. In SAT, from class III obese patients, mRNA and protein expression of BCAA catabolic enzymes and BCKA dehydrogenase (BCKDH) enzyme activity was decreased. Within AA, a concomitant increase in mRNA levels of BCAA metabolizing enzymes was observed, independent of changes in BCKDH protein expression or activity. BMI, indices of tissue dysfunction and duration of hospital stay following surgery correlated with BCAA metabolizing enzyme expression and metabolite levels in AA and SAT. Conclusion: This study proposes that in a setting of obesity, dysregulated BCAA catabolism could be an effective surrogate to determine cardiac surgery outcomes and plausibly predict premature re-hospitalization.

Prognostic factors for changes in the timed 4-stair climb in patients with Duchenne muscular dystrophy, and implications for measuring drug efficacy: A multi-institutional collaboration.

The timed 4-stair climb (4SC) assessment has been used to measure function in Duchenne muscular dystrophy (DMD) practice and research. We sought to identify prognostic factors for changes in 4SC, assess their consistency across data sources, and the extent to which prognostic scores could be useful in DMD clinical trial design and analysis. Data from patients with DMD in the placebo arm of a phase 3 trial (Tadalafil DMD trial) and two real-world sources (Universitaire Ziekenhuizen, Leuven, Belgium [Leuven] and Cincinnati Children's Hospital Medical Center [CCHMC]) were analyzed. One-year changes in 4SC completion time and velocity (stairs/second) were analyzed. Prognostic models included age, height, weight, steroid use, and multiple timed function tests and were developed using multivariable regression, separately in each data source. Simulations were used to quantify impacts on trial sample size requirements. Data on 1-year changes in 4SC were available from the Tadalafil DMD trial (n = 92) Leuven (n = 67), and CCHMC (n = 212). Models incorporating multiple timed function tests, height, and weight significantly improved prognostic accuracy for 1-year change in 4SC (R2: 29%-36% for 4SC velocity, and 29%-34% for 4SC time) compared to models including only age, baseline 4SC and steroid duration (R2:8%-17% for 4SC velocity and 2%-13% for 4SC time). Measures of walking and rising ability contributed important prognostic information for changes in 4SC. In a randomized trial with equal allocation to treatment and placebo, adjustment for such a prognostic score would enable detection (at 80% power) of a treatment effect of 0.25 stairs/second with 100-120 patients, compared to 170-190 patients without prognostic score adjustment. Combining measures of ambulatory function doubled prognostic accuracy for 1-year changes in 4SC completion time and velocity. Randomized clinical trials incorporating a validated prognostic score could reduce sample size requirements by approximately 40%. Knowledge of important prognostic factors can also inform adjusted comparisons to external controls.

Lysophosphatidic acid receptor mRNA levels in heart and white adipose tissue are associated with obesity in mice and humans.

BACKGROUND: Lysophosphatidic acid (LPA) receptor signaling has been implicated in cardiovascular and obesity-related metabolic disease. However, the distribution and regulation of LPA receptors in the myocardium and adipose tissue remain unclear. OBJECTIVES: This study aimed to characterize the mRNA expression of LPA receptors (LPA1-6) in the murine and human myocardium and adipose tissue, and its regulation in response to obesity. METHODS: LPA receptor mRNA levels were determined by qPCR in i) heart ventricles, isolated cardiomyocytes, and perigonadal adipose tissue from chow or high fat-high sucrose (HFHS)-fed male C57BL/6 mice, ii) 3T3-L1 adipocytes and HL-1 cardiomyocytes under conditions mimicking gluco/lipotoxicity, and iii) human atrial and subcutaneous adipose tissue from non-obese, pre-obese, and obese cardiac surgery patients. RESULTS: LPA1-6 were expressed in myocardium and white adipose tissue from mice and humans, except for LPA3, which was undetectable in murine adipocytes and human adipose tissue. Obesity was associated with increased LPA4, LPA5 and/or LPA6 levels in mice ventricles and cardiomyocytes, HL-1 cells exposed to high palmitate, and human atrial tissue. LPA4 and LPA5 mRNA levels in human atrial tissue correlated with measures of obesity. LPA5 mRNA levels were increased in HFHS-fed mice and insulin resistant adipocytes, yet were reduced in adipose tissue from obese patients. LPA4, LPA5, and LPA6 mRNA levels in human adipose tissue were negatively associated with measures of obesity and cardiac surgery outcomes. This study suggests that obesity leads to marked changes in LPA receptor expression in the murine and human heart and white adipose tissue that may alter LPA receptor signaling during obesity.

Enzyme-Mediated Modification of Single-Domain Antibodies for Imaging Modalities with Different Characteristics.

Antibodies are currently the fastest-growing class of therapeutics. Although naked antibodies have proven valuable as pharmaceutical agents, they have some limitations, such as low tissue penetration and a long circulatory half-life. They have been conjugated to toxic payloads, PEGs, or radioisotopes to increase and optimize their therapeutic efficacy. Although nonspecific conjugation is suitable for most in vitro applications, it has become evident that site specifically modified antibodies may have advantages for in vivo applications. Herein we describe a novel approach in which the antibody fragment is tagged with two handles: one for the introduction of a fluorophore or (18)F isotope, and the second for further modification of the fragment with a PEG moiety or a second antibody fragment to tune its circulatory half-life or its avidity. Such constructs, which recognize Class II MHC products and CD11b, showed high avidity and specificity. They were used to image cancers and could detect small tumors.