Multi-head attention-based U-Nets for predicting protein domain boundaries using 1D sequence features and 2D distance maps.

The information about the domain architecture of proteins is useful for studying protein structure and function. However, accurate prediction of protein domain boundaries (i.e., sequence regions separating two domains) from sequence remains a significant challenge. In this work, we develop a deep learning method based on multi-head U-Nets (called DistDom) to predict protein domain boundaries utilizing 1D sequence features and predicted 2D inter-residue distance map as input. The 1D features contain the evolutionary and physicochemical information of protein sequences, whereas the 2D distance map includes the structural information of proteins that was rarely used in domain boundary prediction before. The 1D and 2D features are processed by the 1D and 2D U-Nets respectively to generate hidden features. The hidden features are then used by the multi-head attention to predict the probability of each residue of a protein being in a domain boundary, leveraging both local and global information in the features. The residue-level domain boundary predictions can be used to classify proteins as single-domain or multi-domain proteins. It classifies the CASP14 single-domain and multi-domain targets at the accuracy of 75.9%, 13.28% more accurate than the state-of-the-art method. Tested on the CASP14 multi-domain protein targets with expert annotated domain boundaries, the average per-target F1 measure score of the domain boundary prediction by DistDom is 0.263, 29.56% higher than the state-of-the-art method.

The impact of socioeconomic status on 30-day mortality in hospitalized patients with COVID-19 infection.

Socioeconomic status (SES) impacts outcome in a number of diseases. Our aim was to compare the outcome of hospitalized coronavirus disease 219 (COVID-19) patients in low and high SES group. Prospective cohort study of hospitalized patients with confirmed COVID-19 in three acute hospitals. Electronic case notes were analyzed for baseline characteristics and admission investigations. Scottish index for multiple deprivation (SIMD) was used to divide patients into two groups: more deprived (SIMD 1-5) and less deprived (SIMD 6-10) and results compared. Poor outcome was defined as either need for intubation and/or death. One hundred and seventy-three patients were identified, one was excluded. One hundred and eight (62.8%) were males, mean age was 68.5 ± 14.7 years. Commonest comorbidity was hypertension 87 (50.6%). One hundred and seventeen (68.0%) patients were in more deprived group. Baseline characteristics, admission blood profile and reason for admission were evenly matched in both groups. Outcomes were comparable in both groups: transfer to critical care (27.4% vs 27.3%; P = .991), intubation (18.8% vs 20.2%; P = .853), 30-day all-cause mortality (19.7% vs 14.5%; P = .416) and overall poor outcome (30.8% vs 30.9%; P = .985). Median time to discharge was 7 days longer (17 vs 10 days; P = .018) and median time to death was 4.5 days longer in more deprived group (17 vs 12.5 days; P = .388). Contrary to recent literature on COVID-19 in other geographical areas, our study suggests that the SES does not have any impact on outcome of hospitalized patients with COVID-19, however it negatively impacts length of stay.

Angiotensin converting enzyme inhibitors do not increase the risk of poor outcomes in COVID-19 disease. A multi-centre observational study.

BACKGROUND AND AIMS: Hypertension is associated with an increased risk of severe outcomes with COVID-19 disease. Angiotensin Converting Enzyme (ACE) inhibitors are widely used as a first line medication for the treatment of hypertension in the UK, although their use was suggested in early reports to increase the risk associated with SARS-CoV-2 infection. METHODS: A prospective cohort study of hospitalised patients with laboratory confirmed COVID-19 was conducted across three hospital sites with patients identified on the 9th April 2020. Demographic and other baseline data were extracted from electronic case records, and patients grouped depending on ACE inhibitor usage or not. The 60-day all-cause mortality and need for intubation compared. RESULTS: Of the 173 patients identified, 88 (50.8%) had hypertension. Of these 27 (30.7%) used ACE inhibitors. We did not find significant differences in 60-day all-cause mortality, the requirement for invasive ventilation or length of stay between our patient cohorts after adjusting for covariates. CONCLUSION: This study contributes to the growing evidence supporting the continued use of ACE inhibitors in COVID-19 disease, although adequately powered randomised controlled trials will be needed to confirm effects.

Impact of the COVID-19 pandemic on core surgical training.

BACKGROUND AND AIMS: COVID-19 pandemic has caused significant disruption in training which is even more pronounced in the surgical specialties. We aim to assess the impact of COVID-19 pandemic on core surgical training. METHODS: All core surgical and improving surgical trainees in West of Scotland region were invited to participate in an online voluntary anonymous survey via SurveyMonkey. RESULTS: 28 of 44 (63.6%) trainees responded, 15 (53.6%) were CT1/ST1. 14 (50.0%) working in teaching hospital and 15 (53.6%) working in general surgery. 20 (71.4%) felt that due to the pandemic they have less opportunity to operate as the primary surgeon. 21 (75.0%) have not attended any outpatient clinics. 8 (28.6%) did not have any form of access to the laparoscopic box-trainer. 20 (71.4%) felt their level of confidence in preforming surgical skills has been negatively impacted. 18 (64.3%) found it difficult to demonstrate progress in portfolio. 21 (75.0%) trainees have not attended any teaching. 10 (35.7%) trainees have been off-sick. 8 (28.6%) trainees have felt slightly or significantly more stressed. CONCLUSION: COVID-19 pandemic has an unprecedented negative impact on all aspects of core surgical training. The long term impact on the current cohort of trainees is yet to be seen.

Accurate prediction of protein tertiary structural changes induced by single-site mutations with equivariant graph neural networks.

Predicting the change of protein tertiary structure caused by singlesite mutations is important for studying protein structure, function, and interaction. Even though computational protein structure prediction methods such as AlphaFold can predict the overall tertiary structures of most proteins rather accurately, they are not sensitive enough to accurately predict the structural changes induced by single-site amino acid mutations on proteins. Specialized mutation prediction methods mostly focus on predicting the overall stability or function changes caused by mutations without attempting to predict the exact mutation-induced structural changes, limiting their use in protein mutation study. In this work, we develop the first deep learning method based on equivariant graph neural networks (EGNN) to directly predict the tertiary structural changes caused by single-site mutations and the tertiary structure of any protein mutant from the structure of its wild-type counterpart. The results show that it performs substantially better in predicting the tertiary structures of protein mutants than the widely used protein structure prediction method AlphaFold.

Deep Learning Prediction of Severe Health Risks for Pediatric COVID-19 Patients with a Large Feature Set in 2021 BARDA Data Challenge.

Most children infected with COVID-19 have no or mild symptoms and can recover automatically by themselves, but some pediatric COVID-19 patients need to be hospitalized or even to receive intensive medical care (e.g., invasive mechanical ventilation or cardiovascular support) to recover from the illnesses. Therefore, it is critical to predict the severe health risk that COVID-19 infection poses to children to provide precise and timely medical care for vulnerable pediatric COVID-19 patients. However, predicting the severe health risk for COVID-19 patients including children remains a significant challenge because many underlying medical factors affecting the risk are still largely unknown. In this work, instead of searching for a small number of most useful features to make prediction, we design a novel large-scale bag-of-words like method to represent various medical conditions and measurements of COVID-19 patients. After some simple feature filtering based on logistical regression, the large set of features is used with a deep learning method to predict both the hospitalization risk for COVID-19 infected children and the severe complication risk for the hospitalized pediatric COVID-19 patients. The method was trained and tested the datasets of the Biomedical Advanced Research and Development Authority (BARDA) Pediatric COVID-19 Data Challenge held from Sept. 15 to Dec. 17, 2021. The results show that the approach can rather accurately predict the risk of hospitalization and severe complication for pediatric COVID-19 patients and deep learning is more accurate than other machine learning methods.

High-Performance Deep Learning Toolbox for Genome-Scale Prediction of Protein Structure and Function.

Computational biology is one of many scientific disciplines ripe for innovation and acceleration with the advent of high-performance computing (HPC). In recent years, the field of machine learning has also seen significant benefits from adopting HPC practices. In this work, we present a novel HPC pipeline that incorporates various machine-learning approaches for structure-based functional annotation of proteins on the scale of whole genomes. Our pipeline makes extensive use of deep learning and provides computational insights into best practices for training advanced deep-learning models for high-throughput data such as proteomics data. We showcase methodologies our pipeline currently supports and detail future tasks for our pipeline to envelop, including large-scale sequence comparison using SAdLSA and prediction of protein tertiary structures using AlphaFold2.

Acute pancreatitis following orlistat therapy: report of two cases.

CONTEXT: Orlistat is a pancreatic lipase inhibitor licensed for the treatment of obesity. As obesity rates increase and non-prescription dispensing of orlistat increases, an awareness of its adverse effects is of crucial importance as complications arise more frequently from increased use. Orlistat induced pancreatitis has been described only once previously, but without a diagnostic increase in serum amylase. CASE REPORT: We report the case of two patients who developed severe acute abdominal pain and elevated pancreatic enzymes at 2 and 10 days after starting orlistat. In one case no alterative precipitant was identified. In the other, a predisposing history of pancreatic injury was present. In both cases all other contributory causes were excluded. CONCLUSIONS: Our reports suggest orlistat can trigger drug induced acute pancreatitis in certain patients. For patients presenting with abdominal pain soon after commencing orlistat, a diagnosis of pancreatitis must be considered. We also recommend cautious use of orlistat in patients at risk of pancreatic injury.

Case report: incidental findings of COVID-19 infection on positron emission tomography/computed tomography for staging of a giant gastric gastrointestinal stromal tumor.

We report the incidental finding of COVID-19 in a 59-year-old male, with no significant cardiorespiratory past medical history who underwent a fluorodeoxyglucose positron emission tomography (FDG-PET) scan for investigation of a likely gastric gastrointestinal stromal tumor (GIST). There may be significant discrepancies between clinical symptoms and radiological severity with COVID-19 infection. FDG-PET scanning has the potential to complement traditional radiological imaging in COVID-19 in diagnosis of subclinical diagnosis or early stage disease, as well as monitoring disease progression.

Hartmann's Pouch Stones and Laparoscopic Cholecystectomy: The Challenges and the Solutions.

BACKGROUND & OBJECTIVE: Hartmann's pouch stones (HPS) encountered during laparoscopic cholecystectomy (LC) may hinder safe dissection of the cystic pedicle or be complicated by mucocele, empyema, or Mirizzi syndrome; distorting the anatomy and increasing the risk of bile duct injury. We studied the incidence, presentations, operative challenges, and outcomes of HPS. METHODS: A cohort study of a prospectively maintained database of LCs and bile duct explorations performed by a single surgeon. Patients were divided into two groups: those with HPS and those without. Patients' demographics, clinical presentation, intra-operative findings, and postoperative outcomes were compared. RESULTS: Of the 5136 patients, 612 (11.9%) had HPS. The HPS group were more likely to present with acute cholecystitis (27.9% vs 5.9%, P = .000) and more patients underwent emergency LC (50.7% vs 41.5%, P = .000). The HPS group had more difficult cholecystectomies, with 46.1% vs 11.8% in the non-HPS group being operative difficulty grade 4 and 5. Mucocele, empyema, and Mirizzi syndrome were more common in the HPS group (24.0% vs 3.7% P = .000, 30.9% vs 3.7% P = .000, 1.8% vs 0.9% P = .000, respectively). There was no significant difference in the open conversion rate or complications. CONCLUSION: HPS increase the difficulty of LC. Surgeons should be aware of their presence and should employ appropriate dissection strategies. Sharp or diathermy dissection should be avoided. Dislodging the stone into the gall bladder, stone removal, swab dissection, and cholangiography are useful measures to avoid ductal injury and reduce the conversion rate.

Primary closure of choledochotomy after emergency laparoscopic common bile duct exploration.

INTRODUCTION: The debate into whether or not to drain the common bile duct after laparoscopic common bile duct exploration (LCBDE) for choledocholithiasis continues. Reports are in favour of primary closure of choledochotomy in elective setting. The aim of this study is to evaluate our experience in the primary closure of choledochotomy in emergency setting by analysing indications, operative time, hospital stay, complications, and outcome. METHODS: All patients undergoing LCBDE were prospectively entered into a database incorporating demographic and clinical data including mode of admission, indication for surgery, operation details and clinical outcome. All operations were performed by the same consultant (A. Baker). RESULTS: All 81 patients had primary closure after LCBDE between October 2003 and April 2007. The male to female ratio was 1:3. Median age was 68 years (range 19-90 years). Nineteen (23%) patients had obstructive jaundice, 4 (5%) had cholangitis, 7 (8.6%) had acute pancreatitis, 9 (11.1%) had failed ERCP, and 42 (51.8%) had biliary colic or cholecystitis with positive intraoperative cholangiogram (IOC). Those patients were divided into two groups: elective LCBDE (ElLCBDE) (n=33, 41%) and emergency LCBDE (EmLCBDE) (n=48, 59%). There was no significant difference in the median operative time, median hospital stay and complication rate. Total laparoscopic clearance rate in this series was 95%. CONCLUSIONS: To our knowledge, this is the first publication in the literature in which primary closure after laparoscopic common bile duct exploration in emergency setting was used. With no significant difference in operative time, hospital stay and complications, in experienced hands primary closure of CBD in emergency settings is safe and feasible.

Cleaning the laparoscopic camera port by a microbiology culture swab.

INTRODUCTION: The 5-mm camera is currently widely utilized in Laparoscopic surgery. A disturbing problem for the laparoscopic surgeon is when this Port is becoming smudged by fat or blood, affecting the cleanliness of the 5-mm camera lens after insertion. Due to its narrow, long tube, cleaning this port is tricky. In this paper, we advocate a simple, cheap, and effective method for preventing the staining of the port, using a sterile culture swab stick. METHODS: Once the first port is inserted, it will be cleaned with a sterile culture swab with a repeated in-and-out movement inside the port to clean it circumferentially until there is no further smudging of the swab. The camera is then introduced into the abdominal cavity. Each time either fat or blood stain the port, the culture swab stick is then utilized to wipe it. RESULTS: In our experience, the use of the culture swab stick beforehand to clean the 5-mm camera port prior to insertion renders the use of any other cleaning method unnecessary. CONCLUSION: Using the microbiology sterile culture swab stick to clean the 5-mm camera port is a simple, cheap method to clean the port from any debris.