Computational scoring and experimental evaluation of enzymes generated by neural networks.

In recent years, generative protein sequence models have been developed to sample novel sequences. However, predicting whether generated proteins will fold and function remains challenging. We evaluate a set of 20 diverse computational metrics to assess the quality of enzyme sequences produced by three contrasting generative models: ancestral sequence reconstruction, a generative adversarial network and a protein language model. Focusing on two enzyme families, we expressed and purified over 500 natural and generated sequences with 70-90% identity to the most similar natural sequences to benchmark computational metrics for predicting in vitro enzyme activity. Over three rounds of experiments, we developed a computational filter that improved the rate of experimental success by 50-150%. The proposed metrics and models will drive protein engineering research by serving as a benchmark for generative protein sequence models and helping to select active variants for experimental testing.

Preclinical efficacy of RAF/MEK clamp avutometinib in combination with FAK inhibition in low grade serous ovarian cancer.

OBJECTIVES: Low-grade-serous-ovarian-carcinoma (LGSOC) is characterized by a high recurrence rate and limited therapeutic options. About one-third of LGSOC contains mutations in MAPK pathway genes such as KRAS/NRAS/BRAF. Avutometinib is a dual RAF/MEK inhibitor while defactinib and VS-4718 are focal-adhesion-kinase-inhibitors (FAKi). We determined the preclinical efficacy of avutometinib±VS-4718 in LGSOC patient-derived-tumor-xenografts (PDX). METHODS: Whole-exome-sequencing (WES) was used to evaluate the genetic fingerprint of 3 patient-derived LGSOC (OVA(K)250, PERIT(M)17 and A(PE)148). OVA(K)250 tissue was successfully xenografted as PDX into female CB17/lcrHsd-Prkdc/SCID-mice. Animals were treated with either control, avutometinib, VS-4718, or avutometinib/ VS-4718 once daily five days on and two days off through oral gavage. Mechanistic studies were performed ex vivo using avutometinib±defactinib treated LGSOC tumor samples by western blot. RESULTS: WES results demonstrated wild-type KRAS in all 3 LGSOC. OVA(K)250 PDX showed gain-of-function mutations (GOF) in PTK2 and PTK2B genes, and loss-of-heterozygosity in ADRB2, potentially sensitizing to FAK and RAF/MEK inhibition. The combination of avutometinib/ VS-4718 demonstrated strong tumor-growth inhibition compared to controls starting at day 9 (p < 0.002) in OVA(K)250PDX. By 60 days, mice treated with avutometinib alone and avutometinib/VS-4718 were still alive; compared to median survival of 20 days in control-treated mice and of 35 days in VS-4718-treated mice (p < 0.0001). By western-blot assays exposure of OVA(K)250 to avutometinib, FAKi defactinib and their combination demonstrated decreased phosphorylated FAK (p-FAK) as well as decreased p-ERK. CONCLUSION: Avutometinib, and to a larger extent its combination with FAK inhibitor VS-4718, demonstrated promising in vivo activity against a KRAS wild-type LGSOC-PDX. These data support the ongoing registration-directed study (RAMP201/NCT04625270).

Disassembly of the TRIM56-ATR complex promotes cytoDNA/cGAS/STING axis-dependent intervertebral disc inflammatory degeneration.

As the leading cause of disability worldwide, low back pain (LBP) is recognized as a pivotal socioeconomic challenge to the aging population and is largely attributed to intervertebral disc degeneration (IVDD). Elastic nucleus pulposus (NP) tissue is essential for the maintenance of IVD structural and functional integrity. The accumulation of senescent NP cells with an inflammatory hypersecretory phenotype due to aging and other damaging factors is a distinctive hallmark of IVDD initiation and progression. In this study, we reveal a mechanism of IVDD progression in which aberrant genomic DNA damage promoted NP cell inflammatory senescence via activation of the cyclic GMP-AMP synthase/stimulator of IFN genes (cGAS/STING) axis but not of absent in melanoma 2 (AIM2) inflammasome assembly. Ataxia-telangiectasia-mutated and Rad3-related protein (ATR) deficiency destroyed genomic integrity and led to cytosolic mislocalization of genomic DNA, which acted as a powerful driver of cGAS/STING axis-dependent inflammatory phenotype acquisition during NP cell senescence. Mechanistically, disassembly of the ATR-tripartite motif-containing 56 (ATR-TRIM56) complex with the enzymatic liberation of ubiquitin-specific peptidase 5 (USP5) and TRIM25 drove changes in ATR ubiquitination, with ATR switching from K63- to K48-linked modification, c thereby promoting ubiquitin-proteasome-dependent dynamic instability of ATR protein during NP cell senescence progression. Importantly, an engineered extracellular vesicle-based strategy for delivering ATR-overexpressing plasmid cargo efficiently diminished DNA damage-associated NP cell senescence and substantially mitigated IVDD progression, indicating promising targets and effective approaches to ameliorate the chronic pain and disabling effects of IVDD.

Convolutions are competitive with transformers for protein sequence pretraining.

Pretrained protein sequence language models have been shown to improve the performance of many prediction tasks and are now routinely integrated into bioinformatics tools. However, these models largely rely on the transformer architecture, which scales quadratically with sequence length in both run-time and memory. Therefore, state-of-the-art models have limitations on sequence length. To address this limitation, we investigated whether convolutional neural network (CNN) architectures, which scale linearly with sequence length, could be as effective as transformers in protein language models. With masked language model pretraining, CNNs are competitive with, and occasionally superior to, transformers across downstream applications while maintaining strong performance on sequences longer than those allowed in the current state-of-the-art transformer models. Our work suggests that computational efficiency can be improved without sacrificing performance, simply by using a CNN architecture instead of a transformer, and emphasizes the importance of disentangling pretraining task and model architecture. A record of this paper's transparent peer review process is included in the supplemental information.

Comparison of methodologies for craniofacial soft-tissue cephalometrics: The value of virtual reality.

BACKGROUND: Myriad options are available for plastic surgeons to perform soft-tissue analysis, which is vital to perioperative evaluation and research. Our objective is to compare the accuracy, precision, and efficiency of the available cephalometric modalities for conducting facial soft-tissue measurements. METHODS: Twenty soft-tissue facial measurements were performed by 5 measurers with varying experiences on 5 adult subjects, using 6 methods-manual calipers, cone-beam CT, virtual reality (VR), 3D stereophotogrammetry, iPad-based 3D photogrammetry, and 2-dimensional photographs. Measurement sessions were timed and performed in triplicate, for a total of 9000 measurements. Intraclass correlation coefficient (ICC) was calculated for accuracy and one-way ANOVA was used for comparison. The coefficient of variation (CoV) was compared among groups to evaluate the precision of different methods by considering caliper measurements as the gold standard. RESULTS: ICC among raters was 0.932, indicating excellent reliability. VR was significantly faster than other methods (137 s vs. 217 s for caliper, p < 0.001). CoV was the highest for 2D photographs and the lowest for VR (11.0 vs. 6.4, p < 0.001). The CoV of the caliper was similar to that of other methods, except for 2D photography, which was significantly higher. Measurements with the greatest absolute difference from caliper measurements, across modalities, were those around the eyes (left to right exocanthion), tragion to antitragion, and tragion to exocanthion. CONCLUSION: 2D photography is not an accurate method for cephalometric measurements. VR had the lowest variation between measurements, and was the fastest and equivalent to caliper measurements in accuracy. For studies involving a large number of cephalometrics, VR measurements may be a good option to improve study throughput.

Exploring the Role of an Electrolyte Additive in Suppressing Surface Reconstruction of a Ni-Rich NMC Cathode at Ultrahigh Voltage via Enhanced In Situ and Operando Characterization Methods.

Vinylene carbonate (VC) is a widely used electrolyte additive in lithium-ion batteries for enhanced solid electrolyte interphase formation on the anode side. However, the cathode electrolyte interphase (CEI) formation with VC has received a lot less attention. This study presents a comprehensive investigation employing advanced in situ/operando-based Raman and X-ray absorption spectroscopy (XAS) to explore the effect of electrolyte composition on the CEI formation and suppression of surface reconstruction of LixNiyMnzCo1-y-zO2 (NMC) cathodes. A novel chemical pathway via VC polymerization is proposed based on experimental results. In situ Raman spectra revealed a new peak at 995 cm-1, indicating the presence of C-O semi-carbonates resulting from the radical polymerization of VC. Operando Raman analysis unveiled the formation of NiO at 490 cm-1 in the baseline system under ultrahigh voltage (up to 5.2 V). However, this peak was conspicuously absent in the VC electrolyte, signifying the effectiveness of VC in suppressing surface reconstruction. Further investigation was carried out utilizing in situ XAS compared X-ray absorption near edge structure spectra from cells of 3 and 20 cycles in both electrolytes at different operating voltages. The observed shift at the Ni K-edge confirmed a more substantial reduction of Ni in the baseline electrolyte compared to that in the VC electrolyte, thus indicating less CEI protection in the former. A sophisticated extended X-ray absorption fine structure analysis quantitatively confirmed the effective suppression of rock-salt formation with the VC electrolyte during the charging process, consistent with the operando Raman results. The in situ XAS results thus provided additional support for the key findings of this study, establishing the crucial role of VC polymerization in enhancing CEI stability and mitigating surface reconstruction on NMC cathodes. This work clarifies the relationship between the enhanced CEI layer and NMC degradation and inspires rational electrolyte design for long-cycling NMC cathodes.

Protein structure generation via folding diffusion.

The ability to computationally generate novel yet physically foldable protein structures could lead to new biological discoveries and new treatments targeting yet incurable diseases. Despite recent advances in protein structure prediction, directly generating diverse, novel protein structures from neural networks remains difficult. In this work, we present a diffusion-based generative model that generates protein backbone structures via a procedure inspired by the natural folding process. We describe a protein backbone structure as a sequence of angles capturing the relative orientation of the constituent backbone atoms, and generate structures by denoising from a random, unfolded state towards a stable folded structure. Not only does this mirror how proteins natively twist into energetically favorable conformations, the inherent shift and rotational invariance of this representation crucially alleviates the need for more complex equivariant networks. We train a denoising diffusion probabilistic model with a simple transformer backbone and demonstrate that our resulting model unconditionally generates highly realistic protein structures with complexity and structural patterns akin to those of naturally-occurring proteins. As a useful resource, we release an open-source codebase and trained models for protein structure diffusion.

Persistence of backtracking by human RNA polymerase II.

RNA polymerase II (RNA Pol II) can backtrack during transcription elongation, exposing the 3' end of nascent RNA. Nascent RNA sequencing can approximate the location of backtracking events that are quickly resolved; however, the extent and genome-wide distribution of more persistent backtracking are unknown. Consequently, we developed a method to directly sequence the extruded, "backtracked" 3' RNA. Our data show that RNA Pol II slides backward more than 20 nt in human cells and can persist in this backtracked state. Persistent backtracking mainly occurs where RNA Pol II pauses near promoters and intron-exon junctions and is enriched in genes involved in translation, replication, and development, where gene expression is decreased if these events are unresolved. Histone genes are highly prone to persistent backtracking, and the resolution of such events is likely required for timely expression during cell division. These results demonstrate that persistent backtracking can potentially affect diverse gene expression programs.

Airway pepsinogen A4 identifies lung transplant recipients with microaspiration and predicts chronic lung allograft dysfunction.

BACKGROUND: Aspiration is a known risk factor for adverse outcomes post-lung transplantation. Airway bile acids are the gold-standard biomarker of aspiration; however, they are released into the duodenum and likely reflect concurrent gastrointestinal dysmotility. Previous studies investigating total airway pepsin have found conflicting results on its relationship with adverse outcomes post-lung transplantation. These studies measured total pepsin and pepsinogen in the airways. Certain pepsinogens are constitutively expressed in the lungs, while others, such as pepsinogen A4 (PGA4), are not. We sought to evaluate the utility of measuring airway PGA4 as a biomarker of aspiration and predictor of adverse outcomes in lung transplant recipients (LTRs) early post-transplant. METHODS: Expression of PGA4 was compared to other pepsinogens in lung tissue. Total pepsin and PGA4 were measured in large airway bronchial washings and compared to preexisting markers of aspiration. Two independent cohorts of LTRs were used to assess the relationship between airway PGA4 and chronic lung allograft dysfunction (CLAD). Changes to airway PGA4 after antireflux surgery were assessed in a third cohort of LTRs. RESULTS: PGA4 was expressed in healthy human stomach but not lung. Airway PGA4, but not total pepsin, was associated with aspiration. Airway PGA4 was associated with an increased risk of CLAD in two independent cohorts of LTRs. Antireflux surgery was associated with reduced airway PGA4. CONCLUSIONS: Airway PGA4 is a marker of aspiration that predicts CLAD in LTRs. Measuring PGA4 at surveillance bronchoscopies can help triage high-risk LTRs for anti-reflux surgery.

An Open-Source 3D-Printed Hindlimb Stabilization Apparatus for Reliable Measurement of Stimulation-Evoked Ankle Flexion in Rat.

Currently there are numerous methods to evaluate peripheral nerve stimulation interfaces in rats, with stimulation-evoked ankle torque being one of the most prominent. Commercial rat ankle torque measurement systems and custom one-off solutions have been published in the literature. However, commercial systems are proprietary and costly and do not allow for customization. One-off lab-built systems have required specialized machining expertise, and building plans have previously not been made easily accessible. Here, detailed building plans are provided for a low-cost, open-source, and basic ankle torque measurement system from which additional customization can be made. A hindlimb stabilization apparatus was developed to secure and stabilize a rat's hindlimb, while allowing for simultaneous ankle-isometric torque and lower limb muscle electromyography (EMG). The design was composed mainly of adjustable 3D-printed components to accommodate anatomical differences between rat hindlimbs. Additionally, construction and calibration procedures of the rat hindlimb stabilization apparatus were demonstrated in this study. In vivo torque measurements were reliably acquired and corresponded to increasing stimulation amplitudes. Furthermore, implanted leads used for intramuscular EMG recordings complemented torque measurements and were used as an additional functional measurement in evaluating the performance of a peripheral nerve stimulation interface. In conclusion, an open-source and noninvasive platform, made primarily with 3D-printed components, was constructed for reliable data acquisition of evoked motor activity in rat models. The purpose of this apparatus is to provide researchers a versatile system with adjustable components that can be tailored to meet user-defined experimental requirements when evaluating motor function of the rat hindlimbs.

Resilience of stormwater biofilters following the deposition of wildfire residues: Implication on downstream water quality management in wildfire-prone regions.

Stormwater treatment systems such as biofilters could intercept and remove pollutants from contaminated runoff in wildfire-affected areas, ensuring the protection of water quality downstream. However, the deposition of wildfire residues such as ash and black carbon onto biofilters could potentially impair their stormwater treatment functions. Yet, whether and how wildfire residue deposition could affect biofilter functions is unknown. This study examines the impact of wildfire residue deposition on biofilter infiltration and pollutant removal capacities. Exposure to wildfire residues decreased the infiltration capacity based on the amount of wildfire deposited. Wildfire residues accumulated at the top layer of the biofilter, forming a cake layer, but scraping this layer restored the infiltration capacity. While the deposition of wildfire residues slightly changed the pore water geochemistry, it did not significantly alter the removal of metals and E. coli. Although wildfire residues leached some metals into pore water within the simulated root zone, the leached metals were effectively removed by the compost present in the filter media. Collectively, these results indicate that biofilters downstream of wildfire-prone areas could remain resilient or functional and protect downstream water quality if deposited ash is periodically scraped to restore any loss of infiltration capacity following wildfire residue deposition.

Single-cell transcriptomics of Treg reveals hallmarks and trajectories of immunological aging.

Age-related immunosenescence is characterized by progressive dysfunction of adaptive immune response and increased autoimmunity. Nevertheless, the impact of aging on CD4+ regulatory T cells that are master regulators of the immune system remains largely unclear. Here, we report cellular and molecular hallmarks of regulatory T cells derived from murine lymphoid and adipose tissues at 3, 18, and 24 mo of age, respectively, by analyzing their heterogeneity that displays dynamic changes in transcriptomic effector signatures at a single-cell resolution. Although the proportion of regulatory T cells among total Cd4+ T cells, as well as their expression levels of Foxp3, did not show any global change with time, we have identified 6 transcriptomically distinct clusters of regulatory T cells with cross-tissue conserved hallmarks of aging, including increased numbers of proinflammatory regulatory T cells, reduced precursor cells, increased immature and mature T follicular regulatory cells potentially supported by a metabolic switch from oxidative phosphorylation to glycolysis, a gradual loss of CD150hi regulatory T cells that support hematopoiesis, and increased adipose tissue-specific regulatory T cells that are associated with metabolic disease. To dissect the impact of immunosenescence on humoral immunity, we propose some potential mechanisms underlying T follicular regulatory cell-mediated dysfunction by interactome analysis on T follicular regulatory cells, T follicular helper cells, and B cells during aging. Lastly, spatiotemporal analysis further revealed trajectories of regulatory T-cell aging that demonstrate the most significant changes in marrow and adipose tissues that might contribute to the development of age-related immunosenescence and type 2 diabetes. Taken together, our findings could provide a better understanding of age-associated regulatory T-cell heterogeneity in lymphoid and adipose tissues, as well as regulatory T-cell hallmarks during progressive adaptation to aging that could be therapeutically targeted for rejuvenating the aging immune system in the future.

Needs, barriers and facilitators for a healthier lifestyle in haemodialysis patients: The GoodRENal project.

BACKGROUND: Malnutrition, sedentary lifestyle, cognitive dysfunction and poor psychological well-being are often reported in patients on haemodialysis (HD). AIMS: We aimed to explore needs, barriers and facilitators-as perceived by patients, their carers, and healthcare professionals (HCPs) for increasing the adherence to the diet, to physical activity and cognition and psychological well-being. METHODS: This is an observational cross-sectional study following the STROBE statement. This study is part of an ERASMUS+ project, GoodRENal-aiming to develop digital tools as an educational approach to patients on HD. For that, the GoodRENal comprises HD centers located in four Belgium, Greece, Spain and Sweden. Exploratory questionnaires were developed regarding the perceived needs, barriers and facilitators regarding the diet, physical activity, cognition and psychological well-being from the perspective of patients, their carers and HCPs. RESULTS: In total, 38 patients, 34 carers and 38 HCPs were included. Nutrition: For patients and carers, the main needs to adhere to the diet included learning more about nutrients and minerals. For patients, the main barrier was not being able to eat what they like. Physical activity: As needs it was reported information about type of appropriate physical activity, while fatigue was listed as the main barrier. For Cognitive and emotional state, it was perceived as positive for patients and carers perception but not for HCPs. The HCPs identified as needs working as a team, having access to specialised HCP and being able to talk to patients in private. CONCLUSIONS: Patients and their carers listed as needs guidance regarding nutrition and physical activity but were positive with their cognitive and emotional state. The HCPs corroborated these needs and emphasised the importance of teamwork and expert support.

Trends in past-month cannabis use among US adults across a range of disabilities and health conditions, 2015-2019.

INTRODUCTION: While there is increasing interest in the use of cannabis to manage a range of health-related symptoms, little is known about trends in recent cannabis use with respect to various health conditions. METHODS: We examined data from a US representative sample of noninstitutionalized adults age ≥ 18 from the 2015-2019 National Survey on Drug Use and Health (N = 214,505). We estimated the pooled prevalences followed by linear time trends, overall, and by disability (i.e., difficulty hearing, seeing, thinking, walking, dressing, doing errands) and lifetime (i.e., bronchitis, cancer, diabetes, hepatitis, kidney disease) and current (i.e., asthma, depression, heart disease, hypertension) health condition status using logistic regression. Models with year-by-condition status interaction terms were used to assess differential time trends, adjusting for demographic characteristics. RESULTS: From 2015 to 2019, cannabis use increased significantly among adults with and without each disability and health condition examined. However, the increase was more rapid among those with (versus without) difficulty hearing (89.8% increase [4.9% to 9.3%] vs. 37.9% increase [8.7% to 12.0%], p = 0.015), difficulty walking (84.1% increase [6.3% to 11.6%] vs. 36.8% increase [8.7% to 11.9%], p < 0.001), 2-3 impairments (75.3% increase [9.3% to 16.3%] vs. 36.6% increase [8.2% to 11.2%], p = 0.041), and kidney disease (135.3% increase [3.4% to 8.0%] vs. 38.4% increase [8.6% to 11.9%], p = 0.045). CONCLUSION: Given the potential adverse effects of cannabis, prevention and harm reduction efforts should focus on groups at increasingly higher risk for use, including those with disabilities and kidney disease.

Prevalence and Correlates of Past Year Ecstasy/MDMA Use in the United States.

OBJECTIVES: 3,4-Methylenedioxymethamphetamine (MDMA) (also known as "ecstasy" or "Molly") has regained attention in recent years for its efficacy in treating posttraumatic stress disorder, and the drug was granted breakthrough therapy designation for such use by the US Food and Drug Administration in 2017. However, little is known about the current epidemiology of recreational ecstasy/MDMA use. METHODS: We estimated past-year prevalence and correlates of ecstasy/MDMA use based on a representative sample of noninstitutionalized US individuals 12 years or older from the 2015-2020 National Survey on Drug Use and Health (N = 315,661). RESULTS: An estimated 0.9% (95% confidence interval [CI] = 0.9-1.0) of individuals used ecstasy/MDMA in the past year. Compared with those ages 35-49 years, all younger age groups were at increased odds for use, while those older than 50 years (adjusted odds ratio [aOR] = 0.14, 95% CI = 0.08-0.23) were at low odds for use. Compared with heterosexual men, those identifying as bisexual women (aOR = 1.32, 95% CI = 1.02-1.72) were at increased odds for use, and compared with White individuals, those identifying as Asian (aOR = 1.92, 95% CI = 1.42-2.59), Black (aOR = 1.70, 95% CI = 1.41-2.06), or multiracial (aOR = 1.61, 95% CI = 1.19-2.16) were at increased odds for use. Past-year use of other drugs (e.g., cannabis, ketamine), prescription drug misuse (e.g., pain relievers, stimulants), nicotine dependence (aOR = 1.21, 95% CI = 1.00-1.45), and alcohol use disorder (aOR = 1.41, 95% CI = 1.25-1.58) were also associated with increased odds for use. CONCLUSIONS: While use of ecstasy/MDMA continues to be relatively rare, findings from this study can help inform prevention and harm reduction strategies, especially among certain subpopulations that are at high risk for use.