Achievement of Target Gain Larger than Unity in an Inertial Fusion Experiment

On December 5, 2022, an indirect drive fusion implosion on the National Ignition Facility (NIF) achieved a target gain G_{target} of 1.5. This is the first laboratory demonstration of exceeding "scientific breakeven" (or G_{target}>1) where 2.05 MJ of 351 nm laser light produced 3.1 MJ of total fusion yield, a result which significantly exceeds the Lawson criterion for fusion ignition as reported in a previous NIF implosion [H. Abu-Shawareb et al. (Indirect Drive ICF Collaboration), Phys. Rev. Lett. 129, 075001 (2022)PRLTAO0031-900710.1103/PhysRevLett.129.075001]. This achievement is the culmination of more than five decades of research and gives proof that laboratory fusion, based on fundamental physics principles, is possible. This Letter reports on the target, laser, design, and experimental advancements that led to this result.

AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences

The AlphaFold Database Protein Structure Database (AlphaFold DB, https://alphafold.ebi.ac.uk) has significantly impacted structural biology by amassing over 214 million predicted protein structures, expanding from the initial 300k structures released in 2021. Enabled by the groundbreaking AlphaFold2 artificial intelligence (AI) system, the predictions archived in AlphaFold DB have been integrated into primary data resources such as PDB, UniProt, Ensembl, InterPro and MobiDB. Our manuscript details subsequent enhancements in data archiving, covering successive releases encompassing model organisms, global health proteomes, Swiss-Prot integration, and a host of curated protein datasets. We detail the data access mechanisms of AlphaFold DB, from direct file access via FTP to advanced queries using Google Cloud Public Datasets and the programmatic access endpoints of the database. We also discuss the improvements and services added since its initial release, including enhancements to the Predicted Aligned Error viewer, customisation options for the 3D viewer, and improvements in the search engine of AlphaFold DB.

Evaluation of Calibration Approaches for Indoor Deployments of PurpleAir Monitors

Low-cost air quality monitors are growing in popularity among both researchers and community members to understand variability in pollutant concentrations. Several studies have produced calibration approaches for these sensors for ambient air. These calibrations have been shown to depend primarily on relative humidity, particle size distribution, and particle composition, which may be different in indoor environments. However, despite the fact that most people spend the majority of their time indoors, little is known about the accuracy of commonly used devices indoors. This stems from the fact that calibration data for sensors operating in indoor environments are rare. In this study, we sought to evaluate the accuracy of the raw data from PurpleAir fine particulate matter monitors and for published calibration approaches that vary in complexity, ranging from simply applying linear corrections to those requiring co-locating a filter sample for correction with a gravimetric concentration during a baseline visit. Our data includes PurpleAir devices that were co-located in each home with a gravimetric sample for 1-week periods (265 samples from 151 homes). Weekly-averaged gravimetric concentrations ranged between the limit of detection (3 μg/m3) and 330 μg/m3. We found a strong correlation between the PurpleAir monitor and the gravimetric concentration (R>0.91) using internal calibrations provided by the manufacturer. However, the PurpleAir data substantially overestimated indoor concentrations compared to the gravimetric concentration (mean bias error ≥ 23.6 μg/m3 using internal calibrations provided by the manufacturer). Calibrations based on ambient air data maintained high correlations (R ≥ 0.92) and substantially reduced bias (e.g. mean bias error = 10.1 μg/m3 using a US-wide calibration approach). Using a gravimetric sample from a baseline visit to calibrate data for later visits led to an improvement over the internal calibrations, but performed worse than the simpler calibration approaches based on ambient air pollution data. Furthermore, calibrations based on ambient air pollution data performed best when weekly-averaged concentrations did not exceed 30 μg/m3, likely because the majority of the data used to train these models were below this concentration.

Prenatal iodine supplementation and early childhood neurodevelopment: the PoppiE trial - study protocol for a multicentre randomised controlled trial

INTRODUCTION

Observational studies suggest both low and high iodine intakes in pregnancy are associated with poorer neurodevelopmental outcomes in children. This raises concern that current universal iodine supplement recommendations for pregnant women in populations considered to be iodine sufficient may negatively impact child neurodevelopment. We aim to determine the effect of reducing iodine intake from supplements for women who have adequate iodine intake from food on the cognitive development of children at 24 months of age.

METHODS AND ANALYSIS

A multicentre, randomised, controlled, clinician, researcher and participant blinded trial with two parallel groups. Using a hybrid decentralised clinical trial model, 754 women (377 per group) less than 13 weeks' gestation with an iodine intake of ≥165 µg/day from food will be randomised to receive either a low iodine (20 µg/day) multivitamin and mineral supplement or an identical supplement containing 200) µg/day (amount commonly used in prenatal supplements in Australia), from enrolment until delivery. The primary outcome is the developmental quotient of infants at 24 months of age assessed with the Cognitive Scale of the Bayley Scales of Infant Development, fourth edition. Secondary outcomes include infant language and motor development; behavioural and emotional development; maternal and infant clinical outcomes and health service utilisation of children. Cognitive scores will be compared between groups using linear regression, with adjustment for location of enrolment and the treatment effect described as a mean difference with 95% CI.

ETHICS AND DISSEMINATION

Ethical approval has been granted from the Women's and Children's Health Network Research Ethics Committee (HREC/17/WCHN/187). The results of this trial will be presented at scientific conferences and published in peer-reviewed journals.

TRIAL REGISTRATION NUMBER

NCT04586348.

A randomized, double-blind, placebo-controlled, cross-over trial to evaluate the effect of EstroSense® on 2-hydroxyestrone:16α-hydroxyestrone ratio in premenopausal women

OBJECTIVES

Some estrogen metabolites are associated with increased breast cancer risk, while others are protective. Research efforts have focused on modifiable factors, including bioactive compounds found in food or supplements, promoting estrogen profiles with anti-cancer properties. EstroSense^® is a nutraceutical product with bioactive compounds, including Indole-3-carbinol and green-tea catechins, which may favourably affect estrogen profiles. This study was conducted to determine if EstroSense use, compared to placebo, promotes a higher urinary 2-hydroxyestrone:16α-hydroxyestrone ratio (2-OHE₁:16α-OHE₁), a biomarker associated with a lowered risk of breast cancer.

METHODS

A total of 148 premenopausal women were recruited from British Columbia, Canada to participate in a randomized, double-blind, cross-over, multicentre, placebo-controlled study in which women were randomized to a treatment sequence that consisted of either EstroSense^®, followed by placebo or vice-versa. The women were instructed to consume three capsules per day of EstroSense^® or the placebo for three menstrual cycles (∼12 weeks). The primary outcome was the measurement of 2-OHE1:16α-OHE1 in casual samples at baseline and after each treatment phase.

RESULTS

After 12 weeks of intervention, the mean (95% CI) urinary 2-OHE₁:16α-OHE₁ was 4.55 (2.69, 6.42) (p<0.001) higher following EstroSense than placebo adjusted for baseline values.

CONCLUSIONS

EstroSense use led to markedly higher urinary 2-OHE1:16α-OHE1 than the placebo, a biomarker associated with a lower risk of breast cancer.

REGISTRATION

http://clinicaltrials.gov (NCT02385916).

3D-Beacons: decreasing the gap between protein sequences and structures through a federated network of protein structure data resources

While scientists can often infer the biological function of proteins from their 3-dimensional quaternary structures, the gap between the number of known protein sequences and their experimentally determined structures keeps increasing. A potential solution to this problem is presented by ever more sophisticated computational protein modeling approaches. While often powerful on their own, most methods have strengths and weaknesses. Therefore, it benefits researchers to examine models from various model providers and perform comparative analysis to identify what models can best address their specific use cases. To make data from a large array of model providers more easily accessible to the broader scientific community, we established 3D-Beacons, a collaborative initiative to create a federated network with unified data access mechanisms. The 3D-Beacons Network allows researchers to collate coordinate files and metadata for experimentally determined and theoretical protein models from state-of-the-art and specialist model providers and also from the Protein Data Bank.

Lawson Criterion for Ignition Exceeded in an Inertial Fusion Experiment

For more than half a century, researchers around the world have been engaged in attempts to achieve fusion ignition as a proof of principle of various fusion concepts. Following the Lawson criterion, an ignited plasma is one where the fusion heating power is high enough to overcome all the physical processes that cool the fusion plasma, creating a positive thermodynamic feedback loop with rapidly increasing temperature. In inertially confined fusion, ignition is a state where the fusion plasma can begin "burn propagation" into surrounding cold fuel, enabling the possibility of high energy gain. While "scientific breakeven" (i.e., unity target gain) has not yet been achieved (here target gain is 0.72, 1.37 MJ of fusion for 1.92 MJ of laser energy), this Letter reports the first controlled fusion experiment, using laser indirect drive, on the National Ignition Facility to produce capsule gain (here 5.8) and reach ignition by nine different formulations of the Lawson criterion.

The Amount of Folic Acid in Weekly Iron-Folic Acid Supplements Should Be Increased From 0.4 to 2,8 mg to Reduce Neural Tube Defects (NTD)

Objectives

Weekly iron–folic acid (IFA) supplements are recommended for all menstruating women in countries where anaemia prevalence is > 20%. Anaemia caused by folate deficiency is low worldwide, and the need to include folic acid is in question. Including folic acid might reduce the risk of a neural tube defect (NTD) should a woman become pregnant. Most weekly supplements contain 0.4 mg folic acid; however, WHO recommends 2.8 mg because it is seven times the daily dose effective in reducing NTDs. There is a reluctance to switch to supplements containing 2.8 mg of folic acid because of a lack of evidence that this dose would prevent NTDs. Our aim was to investigate the effect of two doses of folic acid, compared with placebo, on red blood cell (RBC) folate, a biomarker of NTD risk.

Methods

We conducted a three-arm double-blind efficacy trial in Malaysia. Non-pregnant women (n = 331) were randomized to receive 60 mg iron and either 0, 0.4, or 2.8 mg folic acid once weekly for 16 weeks.

Results

At 16 weeks, women receiving 0.4 mg and 2.8 mg folic acid per week had a higher mean RBC folate than those receiving 0 mg (mean difference (95% CI) 84 (54 to 113) and 355 (316 to 394) nmol/L, respectively). Women receiving 2.8 mg folic acid had a mean RBC folate 271 (234 to 309) nmol/L greater mean RBC folate than those receiving 0.4 mg. Moreover, women in the 2.8 mg group were seven times (RR 7.3, 95% CI 3.9 to 13.7) more likely to achieve an RBC folate > 748 nmol/L, a concentration associated with a low risk of NTD, compared with the 0.4 mg group.

Conclusions

IFA supplements containing 2.8 mg folic acid increases RBC folate more than those containing 0.4 mg. Increased availability and access to the 2.8 mg formulation is needed.

Funding Sources

Nutrition International (Ottawa, Ontario).

No Association Between Riboflavin and Choline Status in a Sample of Canadian and Malaysian Women of Reproductive Age

Objectives

Riboflavin and choline are essential nutrients that are metabolically inter-related and play major roles in one-carbon metabolism. Betaine can donate a methyl group to homocysteine, forming dimethylglycine (DMG) and methionine in the liver and kidney. Betaine is synthesized from choline in two irreversible reactions, the first of which is catalyzed by flavin adenine dinucleotide (FAD), a cofactor formed from riboflavin. In animal models, riboflavin status has been shown to influence choline and its metabolites betaine and DMG but there are no human studies. Here we examine whether riboflavin status modifies plasma choline, betaine, and DMG concentrations in healthy women (19–45 y).

Methods

Fasting blood was collected from 206 Canadian and 210 Malaysian women between 2015 and 2016. Riboflavin status was assessed using a functional biomarker, erythrocyte glutathione reductase activity coefficient (EGRac). Plasma choline, betaine, and DMG were determined by LCLC MS/MS. Plasma folate and vitamin B12 were determined using the microbiological method and an immune assay, respectively. General linear models were used to assess the independent relationship between EGRac and each of the choline metabolites with adjustment for potential confounders.

Results

71% of Malaysian women had EGRac ≥1.40 compared to 40% of Canadian women. Betaine, DMG, and vitamin B12 concentrations were significantly higher among Malaysian women compared to Canadian women (40.6 vs. 37.0 mmol/L, 2.7 vs. 2.4 mmol/L and 360 vs. 307 pmol/L, respectively). There were no significant associations between EGRac or riboflavin deficiency (defined as EGRac ≥1.40) and choline or its’ metabolites after adjustments for age, ethnicity, body mass index, plasma folate, and vitamin B12. In the adjusted models, plasma choline was positively associated with vitamin B12 concentrations (B = 0.002, 95% CI: 0.000, 0.003) and plasma betaine was positively associated with plasma folate (B = 0.18, 95%CI: 0.07, 0.29) and vitamin B12 (B = 0.011, 95%CI: 0.002, 0.020).

Conclusions

Overall, riboflavin status was not associated with choline and its metabolites in Canadian and Malaysian women.

Funding Sources

Dairy Farmers of Canada, Saudi Arabian Cultural Bureau in Canada, and BC Children's Hospital Research Institute.

The Effect of Iron Supplements on the Gut Microbiome of Non-pregnant Women of Reproductive Age: A Randomized Controlled Trial

Objectives

Blanket iron supplementation during pregnancy is not recommended in Australia. However, many women take iron as part of a prenatal supplement, often at doses >60 mg/day. Iron supplementation may negatively affect the maternal intestinal microbiome by increasing the abundance of pathogenic bacteria and decreasing commensal bacteria. It is not known if iron supplements alter the microbiome of pregnant women. Here we aim to determine if iron supplementation alters the gut microbiome of non-pregnant women of childbearing age.

Methods

In this two-arm parallel design, double-blind, randomized controlled trial, women (18–45 y) were randomized to take an iron supplement (60 mg elemental iron) or a placebo capsule for 21 days. The primary outcome was microbiota beta-diversity (paired-sample weighted UniFrac dissimilarity) between iron and placebo groups in stool samples collected at baseline and 21 days. A secondary outcome was to determine the effect of iron on Escherichia-Shigella genus relative abundance.

Results

Eighty-two women were randomized, 40 and 42 to iron and placebo groups, respectively. Attrition was < 3% (n = 2) and 97% of women took >80% of their study supplements. On Day 21, there was no difference in mean (95% CI) weighted UniFrac between iron and placebo groups [0.003 (−1.4, 0.7) P = 0.52]. There was no difference in the relative abundance of Escherichia-Shigella after 21 days between groups.

Conclusions

For the outcomes we assessed, we did not find an effect of iron supplementation on the microbiome in non-pregnant women. Studies should be carried out in pregnant women, particularly in settings where water and sanitation are poor, and the natural abundance of pathogenic bacteria is higher.

Funding Sources

South Australian Health and Medical Research Institute, Seed Funding Grants for Early/Mid-Career Researchers.

A follow-up on quantitative and qualitative olfactory dysfunction and other symptoms in patients recovering from COVID-19 smell loss

BACKGROUND

Sudden smell loss is a specific early symptom of COVID-19, which, prior to the emergence of Omicron, had estimated prevalence of ~40% to 75%. Chemosensory impairments affect physical and mental health, and dietary behavior. Thus, it is critical to understand the rate and time course of smell recovery. The aim of this cohort study was to characterize smell function and recovery up to 11 months post COVID-19 infection.

METHODS

This longitudinal survey of individuals suffering COVID-19-related smell loss assessed disease symptoms and gustatory and olfactory function. Participants (n=12,313) who completed an initial survey (S1) about respiratory symptoms, chemosensory function and COVID-19 diagnosis between April and September 2020, were invited to complete a follow-up survey (S2). Between September 2020 and February 2021, 27.5% participants responded (n=3,386), with 1,468 being diagnosed with COVID-19 and suffering co-occurring smell and taste loss at the beginning of their illness.

RESULTS

At follow-up (median time since COVID-19 onset ~200 days), ~60% of women and ~48% of men reported less than 80% of their pre-illness smell ability. Taste typically recovered faster than smell, and taste loss rarely persisted if smell recovered. Prevalence of parosmia and phantosmia was ~10% of participants in S1 and increased substantially in S2: ~47% for parosmia and ~25% for phantosmia. Persistent smell impairment was associated with more symptoms overall, suggesting it may be a key marker of long-COVID illness. The ability to smell during COVID-19 was rated slightly lower by those who did not eventually recover their pre-illness ability to smell at S2.

CONCLUSIONS

While smell ability improves for many individuals who lost it during acute COVID-19, the prevalence of parosmia and phantosmia increases substantially over time. Olfactory dysfunction is associated with broader persistent symptoms of COVID-19, and may last for many months following acute COVID-19. Taste loss in the absence of smell loss is rare. Persistent qualitative smell symptoms are emerging as common long-term sequelae; more research into treatment options is strongly warranted given that even conservative estimates suggest millions of individuals may experience parosmia following COVID-19. Healthcare providers worldwide need to be prepared to treat post COVID-19 secondary effects on physical and mental health.

Speech recognition with a hearing-aid processing scheme combining beamforming with mask-informed speech enhancement

A signal processing approach combining beamforming with mask-informed speech enhancement was assessed by measuring sentence recognition in listeners with mild-to-moderate hearing impairment in adverse listening conditions that simulated the output of behind-the-ear hearing aids in a noisy classroom. Two types of beamforming were compared: binaural, with the two microphones of each aid treated as a single array, and bilateral, where independent left and right beamformers were derived. Binaural beamforming produces a narrower beam, maximising improvement in signal-to-noise ratio (SNR), but eliminates the spatial diversity that is preserved in bilateral beamforming. Each beamformer type was optimised for the true target position and implemented with and without additional speech enhancement in which spectral features extracted from the beamformer output were passed to a deep neural network trained to identify time-frequency regions dominated by target speech. Additional conditions comprising binaural beamforming combined with speech enhancement implemented using Wiener filtering or modulation-domain Kalman filtering were tested in normally-hearing (NH) listeners. Both beamformer types gave substantial improvements relative to no processing, with significantly greater benefit for binaural beamforming. Performance with additional mask-informed enhancement was poorer than with beamforming alone, for both beamformer types and both listener groups. In NH listeners the addition of mask-informed enhancement produced significantly poorer performance than both other forms of enhancement, neither of which differed from the beamformer alone. In summary, the additional improvement in SNR provided by binaural beamforming appeared to outweigh loss of spatial information, while speech understanding was not further improved by the mask-informed enhancement method implemented here.

AlphaFold Protein Structure Database: massively expanding the structural coverage of protein-sequence space with high-accuracy models

The AlphaFold Protein Structure Database (AlphaFold DB, https://alphafold.ebi.ac.uk) is an openly accessible, extensive database of high-accuracy protein-structure predictions. Powered by AlphaFold v2.0 of DeepMind, it has enabled an unprecedented expansion of the structural coverage of the known protein-sequence space. AlphaFold DB provides programmatic access to and interactive visualization of predicted atomic coordinates, per-residue and pairwise model-confidence estimates and predicted aligned errors. The initial release of AlphaFold DB contains over 360,000 predicted structures across 21 model-organism proteomes, which will soon be expanded to cover most of the (over 100 million) representative sequences from the UniRef90 data set.

The AlphaFold Protein Structure Database (AlphaFold DB, https://alphafold.ebi.ac.uk) is an extensive, public database of highly accurate protein structure models. The models are the products of AlphaFold2, an Artificial Intelligence algorithm developed by DeepMind. AlphaFold enabled scientists to investigate an unprecedented number of protein structures. The database we describe here provides access to these predicted models and information on their accuracy. The first version of AlphaFold DB contains over 360,000 models of 21 biologically essential species.

ReIMAGINE: a prostate cancer research consortium with added value through its patient and public involvement and engagement

BACKGROUND

ReIMAGINE aims to improve the current prostate specific antigen (PSA)/biopsy risk stratification for prostate cancer (PCa) and develop a new image-based method (with biomarkers) for diagnosing high/low risk PCa in men. ReIMAGINE's varied patient and public involvement (PPI) and engagement (PE) strategy maximises the impact of its scientific output by informing and shaping the different stages of research.

AIMS

Through including the voice of patients and the public, the ReIMAGINE Consortium aims to translate these different perspectives into the design and implementation process. This will improve the overall quality of the research by: reflecting the needs and priorities of patients and the public, ensuring methods and procedures are feasible and appropriate ensuring information is relevant and accessible to those being recruited to the study identifying dissemination channels relevant to patients/the public and developing outputs that are accessible to a lay audience With support from our patient/user groups, the ReIMAGINE Consortium aims to improve our ability to derive prognostic information and allocate men to the most appropriate and effective therapies, using a novel image-based risk stratification with investigation of non-imaging biomarkers.

FINDINGS

We have been working with patients and the public from initiation of the project to ensure that the research is relevant to men and their families. Our PPI Sub-Committee, led by a PCa patient, has been involved in our dissemination strategy, outreach activities, and study design recommendations. For example, the sub-committee have developed a variety of informative videos relevant and accessible to those being recruited, and organised multiple online research engagement events that are accessible to a lay audience. As quoted by one of the study participants, "the more we present the benefits and opportunities to patients and the public, the more research commitment we obtain, and the sooner critical clinical questions such as PCa diagnostics will be addressed".

Applying and improving AlphaFold at CASP14

We describe the operation and improvement of AlphaFold, the system that was entered by the team AlphaFold2 to the “human” category in the 14th Critical Assessment of Protein Structure Prediction (CASP14). The AlphaFold system entered in CASP14 is entirely different to the one entered in CASP13. It used a novel end‐to‐end deep neural network trained to produce protein structures from amino acid sequence, multiple sequence alignments, and homologous proteins. In the assessors' ranking by summed z scores (>2.0), AlphaFold scored 244.0 compared to 90.8 by the next best group. The predictions made by AlphaFold had a median domain GDT_TS of 92.4; this is the first time that this level of average accuracy has been achieved during CASP, especially on the more difficult Free Modeling targets, and represents a significant improvement in the state of the art in protein structure prediction. We reported how AlphaFold was run as a human team during CASP14 and improved such that it now achieves an equivalent level of performance without intervention, opening the door to highly accurate large‐scale structure prediction.

Bridging the age gap in breast cancer: impact of omission of breast cancer surgery in older women with oestrogen receptor-positive early breast cancer on quality-of-life outcomes

BACKGROUND

Primary endocrine therapy may be an alternative treatment for less fit women with oestrogen receptor (ER)-positive breast cancer. This study compared quality-of-life (QoL) outcomes in older women treated with surgery or primary endocrine therapy.

METHODS

This was a multicentre, prospective, observational cohort study of surgery or primary endocrine therapy in women aged over 70 years with operable breast cancer. QoL was assessed using European Organisation for Research and Treatment of cancer QoL questionnaires QLQ-C30, -BR23, and -ELD14, and the EuroQol Five Dimensions 5L score at baseline, 6 weeks, and 6, 12, 18, and 24 months. Propensity score matching was used to adjust for baseline variation in health, fitness, and tumour stage.

RESULTS

The study recruited 3416 women (median age 77 (range 69-102) years) from 56 breast units. Of these, 2979 (87.2 per cent) had ER-positive breast cancer; 2354 women had surgery and 500 received primary endocrine therapy (125 were excluded from analysis due to inadequate data or non-standard therapy). Median follow-up was 52 months. The primary endocrine therapy group was older and less fit. Baseline QoL differed between the groups; the mean(s.d.) QLQ-C30 global health status score was 66.2(21.1) in patients who received primary endocrine therapy versus 77.1(17.8) among those who had surgery plus endocrine therapy. In the unmatched analysis, changes in QoL between 6 weeks and baseline were noted in several domains, but by 24 months most scores had returned to baseline levels. In the matched analysis, major surgery (mastectomy or axillary clearance) had a more pronounced adverse impact than primary endocrine therapy in several domains.

CONCLUSION

Adverse effects on QoL are seen in the first few months after surgery, but by 24 months these have largely resolved. Women considering surgery should be informed of these effects.

Highly accurate protein structure prediction for the human proteome

Protein structures can provide invaluable information, both for reasoning about biological processes and for enabling interventions such as structure-based drug development or targeted mutagenesis. After decades of effort, 17% of the total residues in human protein sequences are covered by an experimentally determined structure1. Here we markedly expand the structural coverage of the proteome by applying the state-of-the-art machine learning method, AlphaFold2, at a scale that covers almost the entire human proteome (98.5% of human proteins). The resulting dataset covers 58% of residues with a confident prediction, of which a subset (36% of all residues) have very high confidence. We introduce several metrics developed by building on the AlphaFold model and use them to interpret the dataset, identifying strong multi-domain predictions as well as regions that are likely to be disordered. Finally, we provide some case studies to illustrate how high-quality predictions could be used to generate biological hypotheses. We are making our predictions freely available to the community and anticipate that routine large-scale and high-accuracy structure prediction will become an important tool that will allow new questions to be addressed from a structural perspective.

Highly accurate protein structure prediction with AlphaFold

Proteins are essential to life, and understanding their structure can facilitate a mechanistic understanding of their function. Through an enormous experimental effort1-4, the structures of around 100,000 unique proteins have been determined5, but this represents a small fraction of the billions of known protein sequences6,7. Structural coverage is bottlenecked by the months to years of painstaking effort required to determine a single protein structure. Accurate computational approaches are needed to address this gap and to enable large-scale structural bioinformatics. Predicting the three-dimensional structure that a protein will adopt based solely on its amino acid sequence-the structure prediction component of the 'protein folding problem'8-has been an important open research problem for more than 50 years9. Despite recent progress10-14, existing methods fall far short of atomic accuracy, especially when no homologous structure is available. Here we provide the first computational method that can regularly predict protein structures with atomic accuracy even in cases in which no similar structure is known. We validated an entirely redesigned version of our neural network-based model, AlphaFold, in the challenging 14th Critical Assessment of protein Structure Prediction (CASP14)15, demonstrating accuracy competitive with experimental structures in a majority of cases and greatly outperforming other methods. Underpinning the latest version of AlphaFold is a novel machine learning approach that incorporates physical and biological knowledge about protein structure, leveraging multi-sequence alignments, into the design of the deep learning algorithm.

Maternal Thiamine Supplementation Promotes Infants’ Language Processing at 24 Weeks

Objectives

In 2005, researchers observed a link between infants’ access to thiamine (vitamin B1) and their subsequent language functioning: infants fed formula that inadvertently lacked thiamine later displayed significant delays in language and motor development. In the context of a double-blind, randomized, controlled trial, the present research provided the first experimental test of a relationship between early life thiamine exposure and infants’ early language processing.

Methods

335 lactating mothers were randomly assigned to receive capsules containing either 0, 1.2, 2.4, or 10 mg of thiamine daily, from 2 weeks postpartum until infants were 24 weeks. We assessed the integrity of 24-week-old infants’ developing system for language processing by measuring the extent to which they displayed attentional enhancement in response to infant-directed speech (IDS) relative to adult-directed speech (ADS). Such IDS-related attentional enhancement implies that infants can differentiate IDS from ADS, which requires underlying skills for analysis of the complex stream of sound that human speech embodies. Of the 335 infants in the study, analyzable data for the IDS preference task were available for 251.

Results

As predicted, a greater-than-chance percentage of all infants displayed enhanced attention to IDS relative to ADS, replicating prior research (one-sample t(250) = 3.06, P = 0.002). The magnitude of the IDS-elicited attentional-enhancement increased significantly in relation to the thiamine dose that lactating mothers received (beta-weight 0.007, t = 2.57, P = 0.011). Notably, however, when supplementation groups were examined separately, only infants whose mothers received the 10 mg daily supplement displayed a statistically significant IDS-related attentional enhancement (10 mg group: one-sample t(63) = 4.14, P = 0.000; all other groups t’s &lt; 1.41, P’s &gt; 0.16).

Conclusions

Taken together, these findings confirm a link between infants’ post-natal access to thiamine and their language processing facility at 24 weeks. The findings both showcase the value of the IDS preference task for monitoring the integrity of infants’ language processing, and underscore the importance of adequate thiamine for ensuring infants’ optimal language development.

Funding Sources

Bill & Melinda Gates Foundation and the New York Academy of Sciences.

Comparing Three Approaches to Salt Intake Assessment Among Lactating Women in Rural Cambodia

Objectives

Monitoring population salt intake is a critical component of implementing salt fortification programs. In Cambodia, salt is being considered as a vehicle for thiamine fortification to prevent infantile beriberi among breastfed infants. However, salt intake among lactating mothers is not known. The gold standard for assessing sodium intake is repeat 24-hr urinary sodium concentrations. This method has logistical barriers, especially in low-resource settings, and other methods have not been trialed in this population. Here we compare three methods of assessing salt intake in lactating Cambodian women: repeat 24-hr urinary sodium concentrations (USC), repeat 12-hr observed weighed intake records (OWIR), and household salt disappearance (HSD).

Methods

Data from trial: NCT03616288. Salt intake was assessed using the three methods in a subsample of lactating women (n = 104)

between 8 and 22 weeks postpartum. Women were asked to collect two 24-hour urine samples within 7 days. Repeat 12-hr OWIR were collected from women. Household salt disappearance was recorded fortnightly, and was divided by the number of household members to estimate individual intakes. Descriptive household salt use was also recorded. Differences in estimated salt intake from each method were compared using a Kruskal-Wallis test.

Results

Mean (95%CI) estimated salt intakes from repeat 24-hr USC, repeat 12-hr OWIR, and HSD were: 9.0 (8.3, 9.8) g/day, 9.1 (7.9, 10.3) g/day, and 10.9 (9.8, 11.9) g/day, respectively. Estimated intakes from HSD were significantly higher than both 24-hr USC (p = 0.009) and 12-hr OWIR (p = 0.002). Estimated intakes from 24-hr USC and 12-hr OWIR were not statistically different (p = 0.6). Salt was being used for purposes other than consumption, such as cleaning fish and vegetables, in 26% of fortnightly visits.

Conclusions

Repeat 24-hr USC and 12-hr OWIR are both acceptable and logistically feasible methods of salt intake assessment among lactating women in rural Cambodia. While HSD is a less resource-intensive approach, this method over-estimated salt intake. Salt being used for purposes other than consumption may contribute to over-estimation using this method.

Funding Sources

Bill & Melinda Gates Foundation, New York Academy of Sciences, Canadian Institutes of Health Research, Research Nova Scotia.

Neurocognitive Benefits of Maternal Thiamine Supplementation for Breastfed Cambodian Infants

Objectives

Women reliant on rice-heavy diets can have inadequate thiamine intakes, placing breastfed infants at risk of thiamine deficiency and, in turn, neurocognitive impairments. We investigated the impact of maternal thiamine supplementation doses on infants’ cognitive, motor, and language development across the first year.

Methods

In this double-blind, four-parallel arm, randomized controlled trial, healthy mothers of exclusively breastfed newborn infants were recruited in Kampong Thom, Cambodia. At 2-weeks postnatal, women (n = 335) were randomized to one of four treatment groups to consume one capsule/day with varying amounts of thiamine for 22 weeks: 0 mg, 1.2 mg, 2.4 mg, and 10 mg. At 2-, 12-, 24- and 52-weeks of age, infants were assessed with the Mullen Scales of Early Learning (MSEL).

Results

Mixed effects modeling suggest that by 6 months of age, the highest maternal thiamine dose (10 mg/day) held significant benefits for infants’ language development, F's (3,659) &gt; 33.2, P's &lt; 0.001, but generally not for motor or visual reception development. Despite having achieved standardized scores on the MSEL that approximated US norms by 6 months, infants showed a significant drop in all cognitive domains following trial completion, indicating that nutritional interventions beyond 6 months may be necessary.

Conclusions

Findings provide the first experimental evidence that thiamine supplementation among lactating mothers at risk of thiamine deficiency protects their infants’ neurocognitive development, with particular benefit to developing language capacities. Results are consistent with studies that report a widening gap in cognitive test scores over time between children from high vs. low-risk contexts. Important questions remain, particularly with respect to the appropriate duration of thiamine supplementation and/or alternate interventions such as mandatory fortification, with potential to build protective stores of thiamine in preconception or pregnancy.

Funding Sources

Bill & Melinda Gates Foundation and the New York Academy of Sciences (Opportunity ID OPP1176128).

Thiamine Status of Supplemented, Lactating Mothers in Rural Cambodia: A Randomized Controlled Trial

Objectives

Thiamine deficiency is a cause of infant morbidity and mortality throughout Southeast and South Asia. Maternal intake influences human milk thiamine concentrations, thus mother's intake must be improved to combat infant deficiency. However, the dose of supplemental thiamine required by lactating mothers is unknown. We aimed to estimate the maternal oral thiamine dose required to optimize milk thiamine concentrations, and to investigate the impact of various doses on thiamine status biomarkers.

Methods

This was a double-blind, four-parallel arm randomized controlled dose-response trial. At 2 weeks postpartum, healthy mothers were randomized to consume one capsule daily for 22 weeks, containing either 0 mg (placebo, n = 83), 1.2 mg (estimated average requirement, n = 86), 2.4 mg (n = 81), or 10 mg (n = 85) thiamine. Human milk total thiamine, whole blood thiamine diphosphate, and erythrocyte transketolase activity coefficient (ETKac) were assessed. An Emax curve, estimated using a non-linear least squares model, was plotted for human milk. Linear mixed-effects models were used to test for differences between treatment groups for milk and blood biomarkers.

Results

A maternal supplemental dose of 2.35 (95% CI 0.58, 7.01) mg/d was estimated to reach 90% of the maximum average human milk total thiamine concentration of 191 µg/L. The mean (SD) milk thiamine concentration was significantly higher in all intervention groups (183 (91), 190 (105), and 206 (89) µg/L, for 1.2, 2.4, and 10 mg, respectively) compared to placebo (153 (85) µg/L; p &lt; 0.0001), and did not differ from each other. Blood biomarkers followed similar group trends, except for infant ETKac, where only the 10 mg (mean [SD]: 1.18 [0.10]) and placebo (1.12 [0.06]) groups differed significantly (p = 0.003).

Conclusions

While an estimated maternal dose of 2.35 (0.58, 7.01) mg/d was required to reach a milk thiamine concentration of 191 µg/L in Emax dose analyses, group comparisons suggest a daily dose of 1.2 mg/d is sufficient to improve maternal biomarkers to levels similar to higher doses (2.4 and 10 mg/d) and consistent with thiamine-replete populations. However, a higher maternal dose of 10 mg/d was required to improve infant ETKac status compared to other dose groups.

Funding Sources

Bill & Melinda Gates Foundation, The New York Academy of Sciences.

Benefits of Maternal Thiamine Supplementation for Enhancing Cambodian Infants’ Social Responsiveness at 24 Weeks

Objectives

Infants in Cambodia are at risk for thiamine deficiency, with long-term negative consequences for neuro-cognitive development. We investigated the extent to which maternal thiamine supplementation enhances infants’ responsiveness to social engagement. Such responsiveness is both indicative of neuro-cognitive development and beneficial for subsequent neuro-cognitive progress.

Methods

335 lactating Cambodian mothers were randomly assigned to receive a capsule containing 0, 1.2, 2.4, or 10 mg of thiamine daily from 2- until 24-weeks postnatal. We assessed infants’ social responsiveness via a new method – the Primary Engagement Task (PET) – when infants were 2-, 12-, and 24-weeks. In the PET, mothers were asked to elicit and sustain a smile from infants as they added and then removed modalities of engagement (eye-contact/facial expression, voice and touch) across six 30-second “epochs.” Subsequent behavioral coding from video-recordings determined changes in infants’ state across the epochs of the PET.

Results

Preliminary analyses based on 103 of the 335 infants in the PET at 24 weeks revealed that infants displayed significantly increased alertness as mothers introduced new modalities of engagement in the PET, and a partial decline in alertness as modalities were removed (linear contrast F(1,102) = 24.37, P = 0.000; quadratic contrast F(1,102) = 7.27, P = 0.008). Although maternal thiamine dosage did not influence these patterns overall (F's &lt; 1.56, P's &gt; 0.20), higher maternal thiamine dosage was associated with infants displaying enhanced alertness when mothers first offered eye-contact in the PET (beta-weight = 0.026; t = 2.01, P = 0.047). This pattern of alerting to maternal eye-contact was statistically significant only in the 10 mg group (paired t(25) = 3.14, P = 0.004).

Conclusions

Infants whose mothers received 10 mg daily thiamine supplementation displayed increased alertness to opportunities to interact socially. These preliminary findings (a) showcase the value of the PET for assessing infants’ responsiveness to caregiver's efforts to engage them socially, and (b) hint that maternal thiamine supplementation promotes infants’ social responsiveness, thereby underscoring the importance of infants’ early access to adequate thiamine for neuro-cognitive thriving.

Funding Sources

Bill & Melinda Gates Foundation & New York Academy of Sciences.

Bridging the age gap in breast cancer: cluster randomized trial of two decision support interventions for older women with operable breast cancer on quality of life, survival, decision quality, and treatment choices

BACKGROUND

Rates of surgery and adjuvant therapy for breast cancer vary widely between breast units. This may contribute to differences in survival. This cluster RCT evaluated the impact of decision support interventions (DESIs) for older women with breast cancer, to ascertain whether DESIs influenced quality of life, survival, decision quality, and treatment choice.

METHODS

A multicentre cluster RCT compared the use of two DESIs against usual care in treatment decision-making in older women (aged at least ≥70 years) with breast cancer. Each DESI comprised an online algorithm, booklet, and brief decision aid to inform choices between surgery plus adjuvant endocrine therapy versus primary endocrine therapy, and adjuvant chemotherapy versus no chemotherapy. The primary outcome was quality of life. Secondary outcomes included decision quality measures, survival, and treatment choice.

RESULTS

A total of 46 breast units were randomized (21 intervention, 25 usual care), recruiting 1339 women (670 intervention, 669 usual care). There was no significant difference in global quality of life at 6 months after the baseline assessment on intention-to-treat analysis (difference -0.20, 95 per cent confidence interval (C.I.) -2.69 to 2.29; P = 0.900). In women offered a choice of primary endocrine therapy versus surgery plus endocrine therapy, knowledge about treatments was greater in the intervention arm (94 versus 74 per cent; P = 0.003). Treatment choice was altered, with a primary endocrine therapy rate among women with oestrogen receptor-positive disease of 21.0 per cent in the intervention versus 15.4 per cent in usual-care sites (difference 5.5 (95 per cent C.I. 1.1 to 10.0) per cent; P = 0.029). The chemotherapy rate was 10.3 per cent at intervention versus 14.8 per cent at usual-care sites (difference -4.5 (C.I. -8.0 to 0) per cent; P = 0.013). Survival was similar in both arms.

CONCLUSION

The use of DESIs in older women increases knowledge of breast cancer treatment options, facilitates shared decision-making, and alters treatment selection. Trial registration numbers: EudraCT 2015-004220-61 (https://eudract.ema.europa.eu/), ISRCTN46099296 (http://www.controlled-trials.com).

Study protocol for a randomised controlled trial evaluating the effect of folic acid supplementation beyond the first trimester on maternal plasma unmetabolised folic acid in late gestation

INTRODUCTION

Taking folic acid containing supplements prior to and during early pregnancy reduces the risk of neural tube defects. Neural tube defects occur prior to 28 days postconception, after which, there is no proven benefit of continuing to take folic acid. However, many women continue to take folic acid containing supplements throughout the pregnancy. At higher intakes, folic acid is not converted to its active form and accumulates in circulation as unmetabolised folic acid (UMFA). Recently, concerns have been raised about possible links between late gestation folic acid supplementation and childhood allergy, metabolic disease and autism spectrum disorders. We aim to determine if removing folic acid from prenatal micronutrient supplements after 12 weeks gestation reduces circulating levels of maternal UMFA at 36 weeks gestation.

METHODS AND ANALYSIS

This is a parallel-design, double-blinded randomised controlled trial. Women ≥12 and <16 weeks' gestation with a singleton pregnancy and able to give informed consent are eligible to participate. Women (n=100; 50 per group) will be randomised to receive either a micronutrient supplement containing 0.8 mg of folic acid or a micronutrient supplement without folic acid daily from enrolment until delivery. The primary outcome is plasma UMFA concentration at 36 weeks gestation. Secondary outcomes include red blood cell folate and total plasma folate concentration. We will assess whether there is a difference in mean UMFA levels at 36 weeks gestation between groups using linear regression with adjustment for baseline UMFA levels and gestational age at trial entry. The treatment effect will be described as a mean difference with 95% CI.

ETHICS AND DISSEMINATION

Ethical approval has been granted from the Women's and Children's Health Network Research Ethics Committee (HREC/19/WCHN/018). The results of this trial will be presented at scientific conferences and published in peer-reviewed journals.

TRIAL REGISTRATION NUMBER

ACTRN12619001511123.

Using non-mydriatic fundus photography to detect fundus pathology in Australian metropolitan emergency departments: A prospective prevalence and diagnostic accuracy study

OBJECTIVE

To determine the prevalence of fundus pathology in metropolitan Australian EDs utilising a non-mydriatic fundus photography screening programme. Secondary objectives include diagnostic accuracy among emergency physicians compared to telehealth ophthalmologist review.

METHODS

Prospective cross-sectional study investigating non-mydriatic fundus photography as a new diagnostic test in two tertiary Australian EDs. Consecutive adult patients were enrolled if they presented with headache, focal neurological deficit, visual disturbance or diastolic BP >120 mmHg. Diagnostic agreement was determined using kappa statistics and sensitivity and specificity using a reference standard consensus ophthalmology review.

RESULTS

A total of 345 consecutive patients were enrolled among whom 56 (16%, 95% confidence interval [CI] 13-21) had urgent fundus pathology. Agreement between emergency physician and ophthalmic assessment of fundus photographs was 74% (kappa = 0.196, P = 0.001). Emergency physicians had 40% sensitivity (95% CI 27-54) and 82% specificity (95% CI 76-86) for detecting urgent pathology on photographs.

CONCLUSIONS

Fundus photography detects a clinically significant proportion of fundus pathology and urgent diagnoses. Telehealth specialist image review is important to detect some important, time-critical illnesses that can be missed in routine care. This offers an accurate alternative to direct ophthalmoscopy that warrants further research in Australian EDs.

Prevalence, Severity, and Type of Anaemia Among Malaysian Women of Reproductive Age

Objectives

Globally, anaemia affects approximately one third of adolescent girls and women of reproductive age. Anaemia can lead to severe health consequences and can negatively impact social and economic development. Our aim was to understand the prevalence, severity and types of anaemia among Malaysian women.

Methods

A total of n = 331 Malaysian women (18–45 y) who were recruited for a trial on the effects of once weekly iron folic acid supplementation on blood folate levels were included in the analyses for this objective. Fasting venous blood was collected and a complete blood count was conducted using an automated haematology analyser.

Results

The mean ± SD age of participants was 19.0 ± 2.3 years. Eighty-nine % of participants were Malay (n = 295); the remainder identified as Chinese, Indian, or of another ethnicity (11%; n = 36). The mean ± SD body mass index (BMI) of participants was 22.4 ± 4.8 kg/m2, with 60% (n = 199) having a normal BMI (BMI between 18.5–24.9 kg/m2). Overall, the mean ± SD haemoglobin (Hb) concentration was 128.5 ± 11.5 g/L, with 17.5% (n = 58) women presenting as anaemic (Hb &lt; 120 g/L). Of the 58 women with anaemia, 59% (n = 40) had mild anaemia (Hb 110–119 g/L), 29% (n = 17) had moderate anaemia (Hb 80–109 g/L), and ∼2% (n = 1) had severe anaemia (Hb &lt; 80 g/L). Among the participants with anaemia, 52% (n = 30) presented with microcytic cells (mean corpuscular volume &lt; 78 fL), and 72% (n = 42) presented with hypochromic cells (mean corpuscular haemoglobin &lt; 27 pg). None of the women with Hb &lt; 120 g/L had macrocytic or hyperchromic cells.

Conclusions

Anaemia prevalence among Malaysian women of reproductive age in our study was lower than previously reported in a national survey (17.5% vs. 25%, respectively). The majority of anaemia was microcytic, which may be an indication of iron deficiency and/or genetic haemoglobin disorders, both of which are known to be prevalent in Southeast Asia.

Funding Sources

Nutrition International.

Are the Nutrient and Textural Properties of Australian Commercial Infant and Toddler Foods Consistent with Infant Feeding Advice?

Objectives

Australian Infant Feeding Guidelines recommend first foods be iron rich with no added sugars and that nutrient poor discretionary foods be avoided. They also encourage exposing infants to a variety of foods and flavours with increasingly complex textures. Here, we compare nutritional and textural properties of commercially available infant and toddler foods with established guidelines.

Methods

Infant and toddler food products available as of August 2019 were identified by searching for ‘baby food’ on retailer websites and by location in the ‘baby food’ section of major supermarket chains (Coles, Woolworths, Aldi and Foodland) in Adelaide, South Australia. Manufacturer websites were cross checked to identify all available products. Nutrient composition of foods was obtained from the nutrient information panel on the product label, manufacturer or retailer websites. In total, 446 products from 17 brands were reviewed, which accounted for over 90% of the market share.

Results

Most products were poor sources of iron, and 80% of first foods were fruit-based. Half of all products were purees in squeeze pouches and one third of all products were discretionary foods. The nutritional content of many products was inconsistent with guidelines, being low in iron, sweet, smooth in consistency, or classified as discretionary.

Conclusions

Reformulation of products is warranted to improve iron content, particularly in mixed main dishes, expand the range of vegetable only foods, and textural variety. Greater regulatory oversight and front of package labelling are needed to better inform parents and caregivers. Frequent consumption of commercial baby foods low in iron may increase risk of iron deficiency. Excessive consumption of purees via squeeze pouches may also have implications for overweight and obesity risk.

Funding Sources

There are no funding sources to declare for this research.

Metal Contaminants Leached by the Lucky Iron Fish When Boiled in Deionized Distilled Water

Objectives

The Lucky Iron Fish® (LIF) is a reusable fish-shaped iron ingot that was developed to address iron deficiency and iron deficiency anemia in adults and young children under 5 years. The primary objective of this study was to determine the amount of iron and metals leached from the LIF.

Methods

LIF produced in Cambodia, Canada, and India were obtained from the LIF company for analysis. Ingots were used for the first time in this study. All equipment was cleaned prior to use. LIF were boiled in deionized distilled water with lemon juice for 10 and 60 minutes. Water samples were sent to a commercial laboratory for analysis.

Results

When boiled for 10 minutes, the Cambodia, Canada, and India LIF leached concentrations of iron greater than 68,600 ug/L. Quantities of heavy metals leached varied depending on where the ingot was manufactured. When boiled for 10 minutes, arsenic concentrations ranged from 2.5 ug/L to 4.7 ug/L in an India and Cambodia LIF. Chromium ranged from 4–80 ug/L in an India and Cambodia LIF, respectively. Similarly, lead leached ranged from 0.3 ug/L in a Cambodia LIF to 6.3 ug/L in an India LIF.

Conclusions

Despite high quantities of iron leached in water, it is unclear whether the quantity of absorbable iron leached from the LIF is enough to make a difference on iron stores of a young child. As research on the efficacy of the LIF has not been studied in this population. In addition, the potential harms of ingesting these heavy metals must be further investigated.

Funding Sources

None.

Improved protein structure prediction using potentials from deep learning

Protein structure prediction can be used to determine the three-dimensional shape of a protein from its amino acid sequence¹. This problem is of fundamental importance as the structure of a protein largely determines its function²; however, protein structures can be difficult to determine experimentally. Considerable progress has recently been made by leveraging genetic information. It is possible to infer which amino acid residues are in contact by analysing covariation in homologous sequences, which aids in the prediction of protein structures³. Here we show that we can train a neural network to make accurate predictions of the distances between pairs of residues, which convey more information about the structure than contact predictions. Using this information, we construct a potential of mean force⁴ that can accurately describe the shape of a protein. We find that the resulting potential can be optimized by a simple gradient descent algorithm to generate structures without complex sampling procedures. The resulting system, named AlphaFold, achieves high accuracy, even for sequences with fewer homologous sequences. In the recent Critical Assessment of Protein Structure Prediction⁵ (CASP13)-a blind assessment of the state of the field-AlphaFold created high-accuracy structures (with template modelling (TM) scores⁶ of 0.7 or higher) for 24 out of 43 free modelling domains, whereas the next best method, which used sampling and contact information, achieved such accuracy for only 14 out of 43 domains. AlphaFold represents a considerable advance in protein-structure prediction. We expect this increased accuracy to enable insights into the function and malfunction of proteins, especially in cases for which no structures for homologous proteins have been experimentally determined⁷.

Human-level performance in 3D multiplayer games with population-based reinforcement learning

Reinforcement learning (RL) has shown great success in increasingly complex single-agent environments and two-player turn-based games. However, the real world contains multiple agents, each learning and acting independently to cooperate and compete with other agents. We used a tournament-style evaluation to demonstrate that an agent can achieve human-level performance in a three-dimensional multiplayer first-person video game, Quake III Arena in Capture the Flag mode, using only pixels and game points scored as input. We used a two-tier optimization process in which a population of independent RL agents are trained concurrently from thousands of parallel matches on randomly generated environments. Each agent learns its own internal reward signal and rich representation of the world. These results indicate the great potential of multiagent reinforcement learning for artificial intelligence research.

Protein structure prediction using multiple deep neural networks in the 13th Critical Assessment of Protein Structure Prediction (CASP13)

We describe AlphaFold, the protein structure prediction system that was entered by the group A7D in CASP13. Submissions were made by three free-modeling (FM) methods which combine the predictions of three neural networks. All three systems were guided by predictions of distances between pairs of residues produced by a neural network. Two systems assembled fragments produced by a generative neural network, one using scores from a network trained to regress GDT_TS. The third system shows that simple gradient descent on a properly constructed potential is able to perform on par with more expensive traditional search techniques and without requiring domain segmentation. In the CASP13 FM assessors' ranking by summed z-scores, this system scored highest with 68.3 vs 48.2 for the next closest group (an average GDT_TS of 61.4). The system produced high-accuracy structures (with GDT_TS scores of 70 or higher) for 11 out of 43 FM domains. Despite not explicitly using template information, the results in the template category were comparable to the best performing template-based methods.

Computer-Based Connected-Text Training of Speech-in-Noise Perception for Cochlear Implant Users

An interactive method for training speech perception in noise was assessed with adult cochlear implant users. The method employed recordings of connected narratives divided into phrases of 4 to 10 words, presented in babble. After each phrase, the listener identified key words from the phrase from among similar sounding foil words. Nine postlingually deafened adult cochlear implant users carried out 12 hr of training over a 4-week period. Training was carried out at home on tablet computers. The primary outcome measure was sentence recognition in babble. Vowel and consonant identification in speech-shaped noise were also assessed, along with digit span in noise, intended as a measure of some important underlying cognitive abilities. Talkers for speech tests were different from those used in training. To control for procedural learning, the test battery was administered repeatedly prior to training. Performance was assessed immediately after training and again after a further 4 weeks during which no training occurred. Sentence recognition in babble improved significantly after training, with an improvement in speech reception threshold of approximately 2 dB, which was maintained at the 4-week follow-up. There was little evidence of improvement in the other measures. It appears that the method has potential as a clinical intervention. However, the underlying sources of improvement and the extent to which benefits generalize to real-world situations remain to be determined.

P1677Elevation of right-sided pressures and right ventricular echocardiographic parameters: predictors of Exercise Limitation in Patients with Implanted Continuous Flow Left Ventricular Assist Devices

Introduction

Left Ventricular Assist Devices (LVAD) improve survival and functional capacity in patients with advanced heart failure (HF). However, there are potential complications.

Purpose

We sought to determine parameters of exercise intolerance in a group of patients with the HeartWare LVAD (HVAD) compared to a group of HF patients.

Methods

This was a single-centre parallel prospective group-study. Briefly, echocardiograms, right heart catheterisation (RHC) and cardiopulmonary exercise tests were performed in forty-two patients admitted for a heart transplant assessment between August2017 and October2018.Of them 20 belonged to the HVAD group and 22 to the HF group.

Results

In our study, HVAD patients had a better exercise capacity than HF patients, although no significant differences were noted (14.0±5.0 ml/kg/min vs 11.3±3.9 ml/kg/min, p=0.06). To determine exercise tolerance, both HVAD and HF groups were subdivided into 2 groups based on the median peak exercise oxygen consumption (peakVO2) for that group. The table shows the comparison between preserved and non-preserved exercise tolerance in HF and HVADpatients. First of all, in the HVADgroup, all resting RHC pressures were significantly lower in the preserved exercise capacity group. However, in HFpatients there were no statistically significant differences between both subgroups in right-sided pressures, but Thermodilution exercise-induced change in cardiac output (ΔCO) and cardiac index (ΔCI) was significantly higher in the patients with preserved exercise tolerance. Secondly, in the HVADgroup the right ventricle was significantly larger in the reduced exercise tolerance subgroup. Moreover, patients with lower peak VO2 had more significant tricuspid regurgitation. Nevertheless, in HFpatients none of the echocardiographic parameters were related to the exercise capacity.

HF HVAD > Median Peak V02 < Median Peak V02 p > Median Peak V02 < Median Peak V02 p Thermodilution CO, l/min: • Rest 4.3±1.0 4.4±1.8 0.82 4.8±0.8 4.2±1.2 0.21 • Exercise 5.6±1.7 4.8±1.8 0.36 7.1±3.2 4.8±0.8 0.05 • ΔCO 1.26±1.0 0.26±0.7 0.02 2.2±2.5 0.4±0.7 0.05 Right Atrium pressure, mmHg 7.0±4.5 6.8±4.10 0.92 4.3±3.2 10.6±6.40 0.02 Mean Pulmonary Artery pressure, mmHg 26.4±12.6 26.5±10.9 0.97 16.8±5.4 30.5±12.5 0.01 Tricuspid Regurgitation, n (%): • None 1 (9) 1 (9) 1 (12) 0 (0) • Mild 7 (64) 8 (73) 7 (88) 4 (44) • Moderate 2 (18) 0 (0) 0 (0) 4 (44) • Severe 1 (9) 2 (18) 0.36 0 (0) 1 (12) 0.03 Right Ventricle Basal Diastolic Diameter, cm 4.0±1.0 4.1±1.0 0.83 3.7±0.5 4.4±0.5 0.02

Conclusion

Right-sided parameters in the echocardiogram and RHC pressures discriminate between preserved and non-preserved exercise capacity in HVADpatients, but not in HFpatients. In these last patients only ΔCO and ΔCI were statistically correlated with peak exercise oxygen consumption

Acknowledgement/Funding

N. Bouzas-Cruz would like to thank the Spanish Society of Cardiology (Sociedad Española de Cardiología), for her research grant and fellowship

Assessing and responding to anxiety and panic in the Emergency Department

Anxiety and panic symptoms are widespread in the general population. The physical manifestations of anxiety and panic commonly account for people presenting to Emergency Departments (EDs). It is therefore important for ED clinicians to be informed of the numerous causes of anxiety and panic and equipped to respond effectively. This paper describes the underlying pathophysiology of the physical symptoms of anxiety and panic and differential diagnoses to consider. Organic conditions that are associated with symptoms of anxiety and panic are highlighted. Brief interventions are tabled for ED clinicians to use when explaining symptoms, and to promote individual self-management.

Development of a joint mental-health and drug health assessment unit and short-stay unit

OBJECTIVES

There is emerging interest in models of care that focus on assessment and brief inpatient treatment (two to three days) including psychiatric emergency care centre units and short-stay units in Australia. We present the development of a functionally integrated Missenden Assessment Unit and six-bed short-stay unit in the new Professor Marie Bashir Centre at Royal Prince Alfred Hospital in inner-city Sydney. The focus was on collaboration between emergency, drug and alcohol and mental-health services in developing the short-stay unit and Missenden Assessment Unit with joint admission and resource use. We outline the models of care and findings from the 2016 evaluation following the initial two years of operation and consider ongoing challenges.

CONCLUSION

The Missenden Assessment Unit provides an alternative point of presentation for mental-health drug and alcohol patients. The short-stay unit provides coordinated, therapeutic interventions. The Missenden Assessment Unit/short-stay unit reduced the burden of presentations to the emergency department while providing the opportunity for training and collaboration. Further refinement of the models of care should occur with policy development and via research.

Thiamine dose response in human milk with supplementation among lactating women in Cambodia: study protocol for a double-blind, four-parallel arm randomised controlled trial

INTRODUCTION

Thiamine (vitamin B1) deficiency remains a concern in Cambodia where women with low thiamine intake produce thiamine-poor milk, putting their breastfed infants at risk of impaired cognitive development and potentially fatal infantile beriberi. Thiamine fortification of salt is a potentially low-cost, passive means of combating thiamine deficiency; however, both the dose of thiamine required to optimise milk thiamine concentrations as well as usual salt intake of lactating women are unknown.

METHODS AND ANALYSIS

In this community-based randomised controlled trial, 320 lactating women from Kampong Thom, Cambodia will be randomised to one of four groups to consume one capsule daily containing 0, 1.2, 2.4 or 10 mg thiamine as thiamine hydrochloride, between 2 and 24 weeks postnatal. The primary objective is to estimate the dose where additional maternal intake of thiamine no longer meaningfully increases infant thiamine diphosphate concentrations 24 weeks postnatally. At 2, 12 and 24 weeks, we will collect sociodemographic, nutrition and health information, a battery of cognitive assessments, maternal (2 and 24 weeks) and infant (24 weeks only) venous blood samples (biomarkers: ThDP and transketolase activity) and human milk samples (also at 4 weeks; biomarker: milk thiamine concentrations). All participants and their families will consume study-provided salt ad libitum throughout the trial, and we will measure salt disappearance each fortnight. Repeat weighed salt intakes and urinary sodium concentrations will be measured among a subset of 100 participants. Parameters of Emax dose-response curves will be estimated using non-linear least squares models with both 'intention to treat' and a secondary 'per-protocol' (capsule compliance ≥80%) analyses.

ETHICS AND DISSEMINATION

Ethical approval was obtained in Cambodia (National Ethics Committee for Health Research 112/250NECHR), Canada (Mount Saint Vincent University Research Ethics Board 2017-141) and the USA (University of Oregon Institutional Review Board 07052018.008). Results will be shared with participants' communities, as well as relevant government and scientific stakeholders via presentations, academic manuscripts and consultations.

TRIAL REGISTRATION NUMBER

NCT03616288.