A Cdk9-PP1 switch regulates the elongation-termination transition of RNA polymerase II.

The end of the RNA polymerase II (Pol II) transcription cycle is strictly regulated to prevent interference between neighbouring genes and to safeguard transcriptome integrity 1 . The accumulation of Pol II downstream of the cleavage and polyadenylation signal can facilitate the recruitment of factors involved in mRNA 3'-end formation and termination 2 , but how this sequence is initiated remains unclear. In a chemical-genetic screen, human protein phosphatase 1 (PP1) isoforms were identified as substrates of positive transcription elongation factor b (P-TEFb), also known as the cyclin-dependent kinase 9 (Cdk9)-cyclin T1 (CycT1) complex 3 . Here we show that Cdk9 and PP1 govern phosphorylation of the conserved elongation factor Spt5 in the fission yeast Schizosaccharomyces pombe. Cdk9 phosphorylates both Spt5 and a negative regulatory site on the PP1 isoform Dis2 4 . Sites targeted by Cdk9 in the Spt5 carboxy-terminal domain can be dephosphorylated by Dis2 in vitro, and dis2 mutations retard Spt5 dephosphorylation after inhibition of Cdk9 in vivo. Chromatin immunoprecipitation and sequencing analysis indicates that Spt5 is dephosphorylated as transcription complexes traverse the cleavage and polyadenylation signal, concomitant with the accumulation of Pol II phosphorylated at residue Ser2 of the carboxy-terminal domain consensus heptad repeat 5 . A conditionally lethal Dis2-inactivating mutation attenuates the drop in Spt5 phosphorylation on chromatin, promotes transcription beyond the normal termination zone (as detected by precision run-on transcription and sequencing 6 ) and is genetically suppressed by the ablation of Cdk9 target sites in Spt5. These results suggest that the transition of Pol II from elongation to termination coincides with a Dis2-dependent reversal of Cdk9 signalling-a switch that is analogous to a Cdk1-PP1 circuit that controls mitotic progression 4 .

Development and Pilot Testing of a Programmatic System for Competency Assessment in US Anesthesiology Residency Training.

BACKGROUND: In 2018, a set of entrustable professional activities (EPAs) and procedural skills assessments were developed for anesthesiology training, but they did not assess all the Accreditation Council for Graduate Medical Education (ACGME) milestones. The aims of this study were to (1) remap the 2018 EPA and procedural skills assessments to the revised ACGME Anesthesiology Milestones 2.0, (2) develop new assessments that combined with the original assessments to create a system of assessment that addresses all level 1 to 4 milestones, and (3) provide evidence for the validity of the assessments. METHODS: Using a modified Delphi process, a panel of anesthesiology education experts remapped the original assessments developed in 2018 to the Anesthesiology Milestones 2.0 and developed new assessments to create a system that assessed all level 1 through 4 milestones. Following a 24-month pilot at 7 institutions, the number of EPA and procedural skill assessments and mean scores were computed at the end of the academic year. Milestone achievement and subcompetency data for assessments from a single institution were compared to scores assigned by the institution's clinical competency committee (CCC). RESULTS: New assessment development, 2 months of testing and feedback, and revisions resulted in 5 new EPAs, 11 nontechnical skills assessments (NTSAs), and 6 objective structured clinical examinations (OSCEs). Combined with the original 20 EPAs and procedural skills assessments, the new system of assessment addresses 99% of level 1 to 4 Anesthesiology Milestones 2.0. During the 24-month pilot, aggregate mean EPA and procedural skill scores significantly increased with year in training. System subcompetency scores correlated significantly with 15 of 23 (65.2%) corresponding CCC scores at a single institution, but 8 correlations (36.4%) were <30.0, illustrating poor correlation. CONCLUSIONS: A panel of experts developed a set of EPAs, procedural skill assessment, NTSAs, and OSCEs to form a programmatic system of assessment for anesthesiology residency training in the United States. The method used to develop and pilot test the assessments, the progression of assessment scores with time in training, and the correlation of assessment scores with CCC scoring of milestone achievement provide evidence for the validity of the assessments.

Patient perspectives on the unwanted effects of multidisciplinary pain management programmes: A qualitative study.

OBJECTIVE: This study aimed to understand the impact of pain management programmes, focusing on the unwanted effects and their influence on patients' long-term use of self-management strategies. DESIGN: Qualitative study. SETTING: Specialist musculoskeletal hospital in North London, England. PARTICIPANTS: Patients with chronic musculoskeletal pain that have completed a pain management programme. INTERVENTION: Multidisciplinary pain management programmes. MAIN MEASURES: Data were collected regarding patients' experiences and unwanted effects from the pain management programme using semi-structured interviews. Data were analysed using thematic analysis. RESULTS: Fourteen participant interviews were included in the analysis (median age 54 years, 12 females). Four themes were generated from the data: Benefits and burdens, Pain management programme and real life, Social support and Healthcare interventions. Unwanted effects included heightened anxiety related to negative interactions with peers, being in a new environment, worries about ability to cope with the programme, social anxiety from being in a group, the strain on families due to participants being away from home and a sense of abandonment at end of the programme. Burdens associated with implementing pain management strategies were identified, including the emotional burden of imposing their self-management on close family and competing demands with time and energy spent on self-management at the expense of work or home commitments. CONCLUSIONS: Pain management programmes have an important role in helping patients to learn how to self-manage chronic pain. Their unwanted effects and the treatment burdens associated with long-term self-management may be an important consideration in improving the longevity of their beneficial effects.

The Relationship Between Age and 30-Day Outcomes Following Unicompartmental Versus Total Knee Arthroplasty.

BACKGROUND: Understanding the short-term complication profile following unicompartmental knee arthroplasty (UKA) versus total knee arthroplasty (TKA) can improve surgical decision-making and patient outcomes. This study aimed to determine if the difference in risk of 30-day morbidity and mortality between UKA and TKA varied based on patient age. METHODS: This retrospective study of a national quality improvement database using data from 2014 to 2020 included 403,342 patients undergoing UKA (n = 12,324) or TKA (n = 391,018). A generalized additive model evaluated nonlinear relationships between primary outcome and predictors (age, procedure, and procedure × age interaction) using a 1:5 UKA to TKA matched sample. Probabilities and odds ratios (95% confidence interval [CI]) estimated the relative risk of complications by age. RESULTS: In the generalized additive model, TKA patients relative to UKA had 1.30 odds (95% CI 1.19 to 1.43, P < 0.001) of 30-day morbidity and mortality. There was a significant nonlinear relationship between age and primary outcome (P = 0.02), such that the odds were lowest at younger ages. They increased slowly until the age of 65 years, when the slope became steeper. The interaction terms for age and procedure were not significant (P = 0.30). The 30-day probability for short-term complications of a 65-, 75-, and 85-year-olds undergoing UKA was 2.1% (95% CI 1.8 to 2.3), 2.4% (95% CI 2.0 to 2.8), and 3.2% (95% CI 2.3 to 4.1), respectively. The probability of a 65-, 75-, and 85-year-old undergoing TKA was 2.9% (95% CI 2.7 to 3.0), 3.6% (95% CI 3.3 to 3.8), and 5.5% (95% CI 4.7 to 6.3), respectively. CONCLUSIONS: Patients undergoing UKA had a quantifiable lower likelihood of morbidity or mortality than TKA at all ages. These data can provide individualized risk for UKA and TKA across the age spectrum and could be helpful in counseling patients regarding their perioperative risk. LEVEL OF EVIDENCE: III (retrospective comparative study).

High-capacity sample multiplexing for single cell chromatin accessibility profiling.

Single-cell chromatin accessibility has emerged as a powerful means of understanding the epigenetic landscape of diverse tissues and cell types, but profiling cells from many independent specimens is challenging and costly. Here we describe a novel approach, sciPlex-ATAC-seq, which uses unmodified DNA oligos as sample-specific nuclear labels, enabling the concurrent profiling of chromatin accessibility within single nuclei from virtually unlimited specimens or experimental conditions. We first demonstrate our method with a chemical epigenomics screen, in which we identify drug-altered distal regulatory sites predictive of compound- and dose-dependent effects on transcription. We then analyze cell type-specific chromatin changes in PBMCs from multiple donors responding to synthetic and allogeneic immune stimulation. We quantify stimulation-altered immune cell compositions and isolate the unique effects of allogeneic stimulation on chromatin accessibility specific to T-lymphocytes. Finally, we observe that impaired global chromatin decondensation often coincides with chemical inhibition of allogeneic T-cell activation.

Machine Learning to Predict Discharge Destination After Total Knee Arthroplasty and Total Hip Arthroplasty.

Discharge destination impacts costs and perioperative planning for primary total knee (TKA) or hip arthroplasty (THA). The purpose of this study was to create a tool to predict discharge destination in contemporary patients. Models were developed using more than 400,000 patients from the National Surgical Quality Improvement Program database. Models were compared with a previously published model using area under the receiver operating characteristic curve (AUC) and decision curve analysis (DCA). AUC on patients with TKA was 0.729 (95% confidence interval [CI]: 0.719 to 0.738) and 0.688 (95% CI: 0.678 to 0.697) using the new and previous models, respectively. AUC on patients with THA was 0.768 (95% CI: 0.758 to 0.778) and 0.726 (95% CI: 0.714 to 0.737) using the new and previous models, respectively. DCA showed substantially improved net clinical benefit. The new models were integrated into a web-based application. This tool enhances clinical decision making for predicting discharge destination following primary TKA and THA. (Journal of Surgical Orthopaedic Advances 32(4):252-258, 2023).

A Machine-Learning Algorithm to Predict the Likelihood of Prolonged Opioid Use Following Arthroscopic Hip Surgery.

PURPOSE: To develop a machine-learning algorithm and clinician-friendly tool predicting the likelihood of prolonged opioid use (>90 days) following hip arthroscopy. METHODS: The Military Data Repository was queried for all adult patients undergoing arthroscopic hip surgery between 2012 and 2017. Demographic, health history, and prescription records were extracted for all included patients. Opioid use was divided into preoperative use (30-365 days before surgery), perioperative use (30 days before surgery through 14 days after surgery), postoperative use (14-90 days after surgery), and prolonged postoperative use (90-365 days after surgery). Six machine-learning algorithms (Naïve Bayes, Gradient Boosting Machine, Extreme Gradient Boosting, Random Forest, Elastic Net Regularization, and artificial neural network) were developed. Area under the receiver operating curve and Brier scores were calculated for each model. Decision curve analysis was applied to assess clinical utility. Local-Interpretable Model-Agnostic Explanations were used to demonstrate factor weights within the selected model. RESULTS: A total of 6,760 patients were included, of whom 2,762 (40.9%) filled at least 1 opioid prescription >90 days after surgery. The artificial neural network model showed superior discrimination and calibration with area under the receiver operating curve = 0.71 (95% confidence interval 0.68-0.74) and Brier score = 0.21 (95% confidence interval 0.20-0.22). Postsurgical opioid use, age, and preoperative opioid use had the most influence on model outcome. Lesser factors included the presence of a psychological comorbidity and strong history of a substance use disorder. CONCLUSIONS: The artificial neural network model shows sufficient validity and discrimination for use in clinical practice. The 5 identified factors (age, preoperative opioid use, postoperative opioid use, presence of a mental health comorbidity, and presence of a preoperative substance use disorder) accurately predict the likelihood of prolonged opioid use following hip arthroscopy. LEVEL OF EVIDENCE: III, retrospective comparative prognostic trial.

Machine Learning to Predict Fascial Dehiscence after Exploratory Laparotomy Surgery.

BACKGROUND: Fascial dehiscence following exploratory laparotomy is associated with significant morbidity and increased mortality. Previously published risk prediction models for fascial dehiscence are dated and limit a surgeon's ability to perform reliable risk assessment intraoperatively. We sought to determine if machine learning can predict fascial dehiscence after exploratory laparotomy. MATERIALS AND METHODS: A retrospective cohort study was conducted of 93,024 patients undergoing exploratory laparotomy from the 2011-2018 ACS NSQIP data files. Data were divided into training (2011-2016, n = 69,969) and temporal validation (2017-2018, n = 23,055) cohorts. A clinical decision support tool was developed using the model generated via machine learning techniques. RESULTS: 1,332 (1.9%) patients in the training cohort and 390 (1.7%) patients in the temporal validation cohort developed fascial dehiscence. The area under the receiver operating characteristic curve was 0.69 (95% CI 0.66 to 0.72) in the validation cohort. Model predictions demonstrated excellent probability calibration. Decision curve analysis calculates net clinical benefit within a threshold range of 0.8%-4.5%. Operative time, surgical site and deep space infections, and body mass index were among the most important features for model predictions. Finally, operative time, sodium level, and hematocrit demonstrated non-linear relationships with predicted risk. CONCLUSION: A clinical decision support tool for predicting fascial dehiscence after exploratory laparotomy was created and validated on a contemporary, national patient cohort using machine learning. The tool calculates net clinical benefit and can be used at the point of care. Some identified risk factor relationships were found to be complex and non-linear, highlighting the ability of some machine learning applications to capture nuanced, patient-specific risk profiles.

Ultrasound Versus Fluoroscopy for Stellate Ganglion Block: A Cadaveric Study.

OBJECTIVE: Post-Traumatic Stress Disorder (PTSD) is a common psychiatric disorder. Recent investigations have demonstrated effectiveness of Stellate Ganglion Blocks (SGB) for reducing symptoms associated with PTSD. Both fluoroscopic guided and ultrasound guided SGB have been described and are regularly used in clinical practice. This study sought to evaluate differences in block performance when comparing fluoroscopic versus ultrasound guided SGB. DESIGN: Cadaveric Pilot Study. SETTING: Academic Research Laboratory. SUBJECTS: Ten Soft-Cured Human Cadavers. METHODS: Ten soft-cured human cadavers were used after being at room temperature for 3 hours. Fluoroscopic and ultrasound guided injections were both performed on each cadaver, randomized to left or right sidedness. In total, 7 mL of omnipaque and methylene blue (5:1) was injected in each side. Injectate spread was assessed by measuring vertebral body spread under fluoroscopy. Successful staining of the sympathetic trunk was assessed under cadaveric dissection, with visualization of the sympathetic trunk stained with methylene blue. RESULTS: Ultrasound guided injections resulted in successful staining in 9 of 10 injections, while 6 of 10 for fluoroscopic guidance (P = .3034). The average spread in the ultrasound group was 4.0 compared with 5.2 for the fluoroscopic group (P =.088). In the four fluoroscopic guided injections which failed to stain, the injection occurred posterior to the prevertebral fascia. In the single ultrasound guided block that failed to stain, the injection was in the carotid sheath. CONCLUSIONS: While there appeared to be a trend favoring ultrasound guidance, no statistical significance was achieved. This was likely due to this being a limited pilot study. Numerous limitations exist in cadaveric studies, and future investigations should be completed to further study this comparison. That said, the use of the SGB may provide significant relief for patients suffering with PTSD.

Predicting Treatment Success with Facet Syndrome: An Algorithm to Predict Lumbar Radiofrequency Ablation Responders in a Military Population.

OBJECTIVE: Radiofrequency ablation (RFA) of the medial branch nerve is a commonly performed procedure for patients with facet syndrome. RFA has previously been demonstrated to provide long-term functional improvement in approximately 50% of patients, including those who had significant pain relief after diagnostic medial branch block. We sought to identify factors associated with success of RFA for facet pain. DESIGN: Active-duty military patients who underwent lumbar RFA (L3, L4, and L5 levels) over a 3-year period were analyzed. Defense and Veterans Pain Rating Scale (DVPRS) and Oswestry Disability Index (ODI) scores were assessed the day of procedure and at the 2-month and 6-month follow-up. These data were analyzed to identify associations between patient demographics, pain, and functional status and patients' improvement after RFA, with a primary outcome of ODI improvement and a secondary outcome of pain reduction. RESULTS: Higher levels of starting functional impairment (starting ODI scores of 42.9 vs. 37.5; P = 0.0304) were associated with a greater likelihood of improvement in functional status 6 months after RFA, and higher starting pain scores (DVPRS pain scores of 6.1 vs. 5.1; P < 0.0001) were associated with a higher likelihood that pain scores would improve 6 months after RFA. A multivariate logistic regression was then used to develop a scoring system to predict improvement after RFA. The scoring system generated a C-statistic of 0.764, with starting ODI, pain scores, and both gender and smoking history as independent variables. CONCLUSIONS: This algorithm compares favorably to that of diagnostic medial branch block in terms of prediction accuracy (C-statistic of 0.764 vs. 0.57), suggesting that its use may improve patient selection in patients who undergo RFA for facet syndrome.

A mixed-methods survey to explore issues with virtual consultations for musculoskeletal care during the COVID-19 pandemic.

OBJECTIVE: To explore orthopaedic and musculoskeletal clinicians' views and experiences of legal, safety, safeguarding and security issues regarding the use of virtual consultations (VC) during the COVID-19 pandemic. A secondary objective was to suggest ways to overcome these issues. METHODS: A mixed method cross-sectional survey was conducted, seeking the views and experiences of orthopaedic and musculoskeletal medically qualified and Allied Health Professionals in the United Kingdom. Descriptive statistical analysis was employed for quantitative data and a qualitative content analysis undertaken for qualitative data. Findings were presented in accordance with the four key issues. RESULTS: Two hundred and ninety professionals (206 physiotherapists, 78 medically qualified professionals, 6 'other' therapists) participated in the survey. Of the 290 participants, 260 (90%) were not using VC prior to the COVID-19 pandemic, 248 respondents (86%) were unsure whether their professional indemnity insurance covered VC, 136 (47%) had considered how they would handle an issue of safeguarding whilst the remainder had not, 126 (43%) had considered what they would do if, during a virtual consultation, a patient suffered an injury (e.g. bang on their head) or a fall (e.g. mechanical or a medical event like syncope) and 158 (54%) reported they felt the current technological solutions are secure in terms of patient data. Qualitative data provided additional context to support the quantitative findings such as validity of indemnification, accuracy of diagnosis and consent using VC, safeguarding issues; and security and sharing of data. Potential changes to practice have been proposed to address these issues. CONCLUSIONS: VC have been rapidly deployed since the onset of the COVID-19 pandemic often without clear guidance or consensus on many important issues. This study identified legal, safeguarding, safety and security issues. There is an urgent need to address these and develop local and national guidance and frameworks to facilitate ongoing safe virtual orthopaedic practice beyond the COVID-19 pandemic.

Nascent RNA sequencing reveals a dynamic global transcriptional response at genes and enhancers to the natural medicinal compound celastrol.

Most studies of responses to transcriptional stimuli measure changes in cellular mRNA concentrations. By sequencing nascent RNA instead, it is possible to detect changes in transcription in minutes rather than hours and thereby distinguish primary from secondary responses to regulatory signals. Here, we describe the use of PRO-seq to characterize the immediate transcriptional response in human cells to celastrol, a compound derived from traditional Chinese medicine that has potent anti-inflammatory, tumor-inhibitory, and obesity-controlling effects. Celastrol is known to elicit a cellular stress response resembling the response to heat shock, but the transcriptional basis of this response remains unclear. Our analysis of PRO-seq data for K562 cells reveals dramatic transcriptional effects soon after celastrol treatment at a broad collection of both coding and noncoding transcription units. This transcriptional response occurred in two major waves, one within 10 min, and a second 40-60 min after treatment. Transcriptional activity was generally repressed by celastrol, but one distinct group of genes, enriched for roles in the heat shock response, displayed strong activation. Using a regression approach, we identified key transcription factors that appear to drive these transcriptional responses, including members of the E2F and RFX families. We also found sequence-based evidence that particular transcription factors drive the activation of enhancers. We observed increased polymerase pausing at both genes and enhancers, suggesting that pause release may be widely inhibited during the celastrol response. Our study demonstrates that a careful analysis of PRO-seq time-course data can disentangle key aspects of a complex transcriptional response, and it provides new insights into the activity of a powerful pharmacological agent.

Divergence of a conserved elongation factor and transcription regulation in budding and fission yeast.

Complex regulation of gene expression in mammals has evolved from simpler eukaryotic systems, yet the mechanistic features of this evolution remain elusive. Here, we compared the transcriptional landscapes of the distantly related budding and fission yeast. We adapted the Precision Run-On sequencing (PRO-seq) approach to map the positions of RNA polymerase active sites genome-wide in Schizosaccharomyces pombe and Saccharomyces cerevisiae. Additionally, we mapped preferred sites of transcription initiation in each organism using PRO-cap. Unexpectedly, we identify a pause in early elongation, specific to S. pombe, that requires the conserved elongation factor subunit Spt4 and resembles promoter-proximal pausing in metazoans. PRO-seq profiles in strains lacking Spt4 reveal globally elevated levels of transcribing RNA Polymerase II (Pol II) within genes in both species. Messenger RNA abundance, however, does not reflect the increases in Pol II density, indicating a global reduction in elongation rate. Together, our results provide the first base-pair resolution map of transcription elongation in S. pombe and identify divergent roles for Spt4 in controlling elongation in budding and fission yeast.

PPP1CC2 can form a kinase/phosphatase complex with the testis-specific proteins TSSK1 and TSKS in the mouse testis.

The mouse protein phosphatase gene Ppp1cc is essential for male fertility, with mutants displaying a failure in spermatogenesis including a widespread loss of post-meiotic germ cells and abnormalities in the mitochondrial sheath. This phenotype is hypothesized to be responsible for the loss of the testis-specific isoform PPP1CC2. To identify PPP1CC2-interacting proteins with a function in spermatogenesis, we carried out GST pull-down assays in mouse testis lysates. Amongst the identified candidate interactors was the testis-specific protein kinase TSSK1, which is also essential for male fertility. Subsequent interaction experiments confirmed the capability of PPP1CC2 to form a complex with TSSK1 mediated by the direct interaction of each with the kinase substrate protein TSKS. Interaction between PPP1CC2 and TSKS is mediated through an RVxF docking motif on the TSKS surface. Phosphoproteomic analysis of the mouse testis identified a novel serine phosphorylation site within the TSKS RVxF motif that appears to negatively regulate binding to PPP1CC2. Immunohistochemical analysis of TSSK1 and TSKS in the Ppp1cc mutant testis showed reduced accumulation to distinct cytoplasmic foci and other abnormalities in their distribution consistent with the loss of germ cells and seminiferous tubule disorganization observed in the Ppp1cc mutant phenotype. A comparison of Ppp1cc and Tssk1/2 knockout phenotypes via electron microscopy revealed similar abnormalities in the morphology of the mitochondrial sheath. These data demonstrate a novel kinase/phosphatase complex in the testis that could play a critical role in the completion of spermatogenesis.

Perspectives of the World Health Organization's physical activity guidelines among patients with musculoskeletal conditions: A mixed-methods survey.

INTRODUCTION: Physical activity (PA) promotion is a core element of musculoskeletal rehabilitation. Many people with musculoskeletal conditions do not meet PA guidelines, such as those provided by the World Health Organization (WHO). This study aimed to explore the level of awareness and perspectives on the WHO PA guidelines among people with musculoskeletal conditions and identify how patients can be supported with PA in clinical practice. METHODS: A mixed-method cross-sectional survey was conducted using a uniquely developed questionnaire, which was disseminated to patients attending outpatient physiotherapy or occupational therapy appointments. The questionnaire collected data on participant characteristics, awareness and perspectives on the WHO guidelines and suggestions for supporting patients with PA in clinical practice. Descriptive statistics and content analysis were used to analyse the quantitative and qualitative data respectively. RESULTS: One-hundred and two responses were included in the analysis. Forty-two percent of respondents were aware of the guidelines, 50% knew why they existed and 53% felt they were achievable. Perspectives varied, ranging from the guidelines being positive and aiding motivation to being unachievable. Some respondents felt an individualised approach is required. Suggestions for supporting patients with PA included facilitating social support, providing education, exercise groups and personalised advice, and signposting to community services. CONCLUSION: There is limited awareness of the WHO PA guidelines and a variety of perspectives on the guidelines among patients with musculoskeletal conditions. There is a need to enhance the promotion of the guidelines in clinical practice, whilst considering local context and individual patient's circumstances.

Fine-Tuning Large Language Models to Enhance Programmatic Assessment in Graduate Medical Education.

Background: Natural language processing is a collection of techniques designed to empower computer systems to comprehend and/or produce human language. The purpose of this investigation was to train several large language models (LLMs) to explore the tradeoff between model complexity and performance while classifying narrative feedback on trainees into the Accreditation Council for Graduate Medical Education subcompetencies. We hypothesized that classification accuracy would increase with model complexity. Methods: The authors fine-tuned several transformer-based LLMs (Bidirectional Encoder Representations from Transformers [BERT]-base, BERT-medium, BERT-small, BERT-mini, BERT-tiny, and SciBERT) to predict Accreditation Council for Graduate Medical Education subcompetencies on a curated dataset of 10 218 feedback comments. Performance was compared with the authors' previous work, which trained a FastText model on the same dataset. Performance metrics included F1 score for global model performance and area under the receiver operating characteristic curve for each competency. Results: No models were superior to FastText. Only BERT-tiny performed worse than FastText. The smallest model with comparable performance to FastText, BERT-mini, was 94% smaller. Area under the receiver operating characteristic curve for each competency was similar on BERT-mini and FastText with the exceptions of Patient Care 7 (Situational Awareness and Crisis Management) and Systems-Based Practice. Discussion: Transformer-based LLMs were fine-tuned to understand anesthesiology graduate medical education language. Complex LLMs did not outperform FastText. However, equivalent performance was achieved with a model that was 94% smaller, which may allow model deployment on personal devices to enhance speed and data privacy. This work advances our understanding of best practices when integrating LLMs into graduate medical education.

Single-cell analysis of chromatin and expression reveals age- and sex-associated alterations in the human heart.

Sex differences and age-related changes in the human heart at the tissue, cell, and molecular level have been well-documented and many may be relevant for cardiovascular disease. However, how molecular programs within individual cell types vary across individuals by age and sex remains poorly characterized. To better understand this variation, we performed single-nucleus combinatorial indexing (sci) ATAC- and RNA-Seq in human heart samples from nine donors. We identify hundreds of differentially expressed genes by age and sex and find epigenetic signatures of variation in ATAC-Seq data in this discovery cohort. We then scale up our single-cell RNA-Seq analysis by combining our data with five recently published single nucleus RNA-Seq datasets of healthy adult hearts. We find variation such as metabolic alterations by sex and immune changes by age in differential expression tests, as well as alterations in abundance of cardiomyocytes by sex and neurons with age. In addition, we compare our adult-derived ATAC-Seq profiles to analogous fetal cell types to identify putative developmental-stage-specific regulatory factors. Finally, we train predictive models of cell-type-specific RNA expression levels utilizing ATAC-Seq profiles to link distal regulatory sequences to promoters, quantifying the predictive value of a simple TF-to-expression regulatory grammar and identifying cell-type-specific TFs. Our analysis represents the largest single-cell analysis of cardiac variation by age and sex to date and provides a resource for further study of healthy cardiac variation and transcriptional regulation at single-cell resolution.

The NELF pausing checkpoint mediates the functional divergence of Cdk9.

Promoter-proximal pausing by RNA Pol II is a rate-determining step in gene transcription that is hypothesized to be a prominent point at which regulatory factors act. The pausing factor NELF is known to induce and stabilize pausing, but not all kinds of pausing are NELF-mediated. Here, we find that NELF-depleted Drosophila melanogaster cells functionally recapitulate the NELF-independent pausing we previously observed in fission yeast (which lack NELF). Critically, only NELF-mediated pausing establishes a strict requirement for Cdk9 kinase activity for the release of paused Pol II into productive elongation. Upon inhibition of Cdk9, cells with NELF efficiently shutdown gene transcription, while in NELF-depleted cells, defective, non-productive transcription continues unabated. By introducing a strict checkpoint for Cdk9, the evolution of NELF was likely critical to enable increased regulation of Cdk9 in higher eukaryotes, as Cdk9 availability can be restricted to limit gene transcription without inducing wasteful, non-productive transcription.