Edinburgh DataShare
Not a member yet
7718 research outputs found
Sort by
Metagenomic characterisation of the caecal microbiota of XCR1 knockout chickens (ANI 99% MAGs: ENA project PRJEB64517)
No full textXCR1+ dendritic cells play an important role in limiting the immune response to members of the gut microbiota, preventing inflammation and maintaining gut homeostasis. Mice lacking XCR1+ cDCs have a proinflammatory gut environment and are prone to opportunistic infections by commensals. As chickens contain greater numbers of these cells in their guts than mammals, it is expected that they would play a larger role in controlling gut inflammation. Increased inflammation caused by a lack of these cells is expected to select against commensal bacteria and promote opportunistic pathogen growth.
Raw sequencing files, primary assemblies and metagenome assembled genomes (MAGs) that were dereplicated at the species level (95% ANI) have been uploaded to the European Nucleotide Archive (project PRJEB64517). This dataset contains all MAGs produced from this project, dereplicated at a strain level (99% ANI).All 670 MAG files are supplied as fa.gz files within MAG_ANI_FASTA_FILES.zip. Checkm2 quality information is also provided for each file: quality_report.tsv. Taxonomic classifications, as produced by gtdb-tk are stored in gtdb_99.ar53.summary.tsv and gtdb_99.bac120.summary.tsv. For metadata on chicken samples from which these MAGs were produced, see ENA project PRJEB64517
SUPERSEDED - Summary of data retrieval process
No full text##This item has been replaced by the one which can be found at https://doi.org/10.7488/ds/7708 ## This file contains a detailed discussion of how we went about corpus construction, with a particular focus on data retrieval. It discusses our use of Nexis and Factiva, Google and GDELT (via API), as well as the web scraping procedures used
FOI data used in Voice of America study
Get PDFThis dataset relates to a research project that explored the increased politicisation of the Voice of America (VOA) international news network during the period when a Trump appointee, Michael Pack, was the CEO of the US Agency for Global Media (USAGM). We analyse why and how VOA was captured by Trump allies, and the effects of this increased politicisation on the US-funded network and on democratic backsliding within the US more broadly.
The research analysed 15,000 pages of material, 12,000 pages of this consisted of private correspondence and internal documents released by USAGM in response to Freedom of Information (FOI) requests. The deposited spreadsheet provides details on the nature and location of these archives, while the methodology document details how we combined analysis of FOI material with semi-structured interviews, the analysis of news and social media, and document analysis of contextual material, including government reports, court verdicts, and documentation released by the January 6 Committee. The dataset relates the academic monograph, titled: Wright, Kate, Scott, Martin, Bunce, Mel. Capturing News, Capturing Democracy: Trump and the Voice of America. (Accepted by Oxford University Press)
Catchment-Averaged Erosion Rates Reveal Signals of Divide Migration and Drainage Capture - Datasets
No full textThis dataset relates to work presented in the article Hoskins A., Attal M., Mudd S., Castillo M., 'Catchment-Averaged Erosion Rates Reveal Signals of Divide Migration and Drainage Capture' (in submission).
Abstract: Divide migration and drainage capture contribute to drainage reorganisation. The relative contributions of each are debated, alongside the extent to which an observable signal of drainage reorganisation may be preserved in quantifiable erosion rates. We numerically model divide migration and drainage capture, and monitor the effects on catchment-averaged erosion rates in the growing (area gaining) and shrinking (area losing) catchments. Divide migration produces a rapid increase in catchment-averaged erosion rates in the headwaters of the growing catchment. However, we find this catchment-averaged erosion rate signal is quickly obscured with increasing distance downstream in non-uniform uplift settings, limiting our ability to detect divide migration through catchment-averaged erosion rate measurements in non-uniform uplift settings. Drainage capture produces the strongest catchment-averaged erosion rate signal immediately adjacent to the point of capture. We find this signal persists in the landscape longest, and without depleting in magnitude, in the area upstream of the point of capture. The Sierra la Laguna mountain range (Mexico) displays substantial evidence of recent and ongoing drainage capture across the main drainage divide, including: beheaded catchments, windgaps, barbed drainages, chi profiles and across divide Gilbert Metrics. We use the Sierra la Laguna to test the detectability of drainage reorganisation related catchment-averaged erosion rate signals in a natural setting. 10Be-derived catchment-averaged erosion rates are found to be twice as fast in the suspected growing catchment headwaters (0.17 mm/yr) relative to the shrinking catchment headwaters (0.09 mm/yr). Catchment-averaged erosion rates, with distance downstream, share similarities with our drainage capture modeling
Feedback forms from public engagement on cattle breeding using Democs card game
No full textThe EC BovReg project on the functional genomics of cattle breeding, used a specially created Democs card game for public engagement to explain cattle breeding and genomics, and to explore ethical questions about breeding priorities and practices, with small groups of publics. Democs is an established public engagement tool to enable groups to explore technical issues through the medium of cards. During the project, a total of 20 Democs games were played by 118 people, primarily across the UK, Finland and the Netherlands (8th July 2021 - 31 August 2023). The main aim of the games was to engage publics with developments in cattle breeding and the factors and ethical issues which bear upon these.
The outputs from the game also give some quantitative and qualitative data, helping us to understand how players expressed their views. Individual responses were invited from players on how important they considered five breeding goals (voting sheet A), questions about ethical limits to selection, on genome editing and on breeding for resilience (voting sheet B) , and how much extra they might be willing to pay for dairy or beef products produced from cattle bred for particular goals (disease resistance, reduced methane, animal health and resilience) (voting sheet C). The relatively small number of players and the self-selecting nature of the groups mean that these data reflect the sample of the people playing the game and cannot simply be extrapolated to wider publics. The willingness to pay extra is unlikely to represent actual values people are willing to pay, rather the question was used as a way of measuring strength of opinion on various applications of cattle breeding.
Games with prefix J, T and JV were played in Finland
Games with prefix U were played in the Netherlands
Games with prefix D and BVGA1 were played by groups in international livestock meetings
Games with prefix S, R, E and H were played in the U
Mismatch between conservation higher education skills training and contemporary conservation needs
No full textData associated with the publication: Slater, H. et al. (2024) Mismatch between conservation higher education skills training and contemporary conservation needs. ### Abstract ### Scholars have detailed the range of skills required for conservation and voiced concerns that training is not fit for purpose. Still, we have little understanding of what skills conservation education aims to teach. This study uses survey data and content analysis of online module descriptions to examine skills and methods teaching in conservation higher education across the UK and Australia. We found most conservation-specific modules aimed to develop disciplinary and communication skills, but fewer than half aimed to develop interpersonal or project management skills. Social science methods training was absent from the core offering of over half of the conservation degrees reviewed. To prepare students for conservation careers and the complex problems they will encounter, the conservation education sector should further focus on building essential non-academic skills and provide training on the breadth of methods that contribute to effective conservation science. This analysis can help educators to reflect on teaching aims and forge a curriculum that will best prepare students for contemporary conservation challenges.See README.txt fil
Official Data Repository for DRED: Zero-Shot Transfer in Reinforcement Learning via Data-Regularised Environment Design
No full textAutonomous agents trained using deep reinforcement learning (RL) often lack the ability to successfully generalise to new environments, even when these environments share characteristics with the ones they have encountered during training. In this work, we investigate how the sampling of individual environment instances, or levels, affects the zero-shot generalisation (ZSG) ability of RL agents. We discover that, for deep actor-critic architectures sharing their base layers, prioritising levels according to their value loss minimises the mutual information between the agent's internal representation and the set of training levels in the generated training data. This provides a novel theoretical justification for the regularisation achieved by certain adaptive sampling strategies. We then turn our attention to unsupervised environment design (UED) methods, which assume control over level generation. We find that existing UED methods can significantly shift the training distribution, which translates to low ZSG performance. To prevent both overfitting and distributional shift, we introduce data-regularised environment design (DRED). DRED generates levels using a generative model trained to approximate the ground truth distribution of an initial set of level parameters. Through its grounding, DRED achieves significant improvements in ZSG over adaptive level sampling strategies and UED methods. Our code and experimental data are available at: Garcin et al, "DRED: Zero-Shot Transfer in Reinforcement Learning via Data-Regularised Environment Design", 2024, (https://github.com/uoe-agents/dred).Experimental data and model repository to reproduce experiments from DRED: Zero-Shot Transfer in Reinforcement Learning via Data-Regularised Environment Design (https://arxiv.org/abs/2402.03479). Plotting code is included to reproduce all figure in the paper.
Structure:
- procgen-mi: contains data relevant to the Procgen experiments.
- dred: contains data relevant to the DRED experiments in MiniGrid.
- dred_pretrained_vae.pt: model weights for the pre-trained VAE used in DRED MiniGrid experiments.
Please refer to the README for our code repository for more detailed information, available at
https://github.com/uoe-agents/dre
genlogistic.ssc: script to estimate theta in the theta-logistic (or generalized logistic) population model from 'On the regulation of populations of mammals, birds, fish, and insects'
No full textThis is version 1.1 of genlogistic.ssc, released in January 2006. genlogistic.ssc is a script written by Daniel Barker and Michael Denham, for estimating theta in the theta-logistic (or generalized logistic) population model. The generalized logistic equation is: pgr = r0[1-(N/K)^theta], where pgr is the population growth rate, N is the population size, K is the carrying capacity, r0 represents pgr at N = 0, and theta describes the curvature of the relationship. The script will estimate theta for one or more population time series provided by the user. genlogistic.ssc accompanies Sibly et al. (2005, Science 309:607-610, doi:10.1126/science.1110760).genlogistic_1_1.zip contains the script itself (genlogistic.ssc; plain text) and an example input file containing hypothetical data, to demonstrate the correct format (genlogistic_in.xls; Microsoft Excel spreadsheet). Instructions are given in a long comment at the start of the genlogistic.ssc script. genlogistic.ssc is an S-Plus script, developed using Insightful S-Plus for Microsoft Windows. However, with slight adaptation it should be possible to make it work under the R software environment instead
Regions where most false positives are found while segmenting perivascular spaces from brain MRI
No full textDickie and colleagues generated seven age-specific brain tissue sets/atlases of T1-weighted brain MRI illustrative of the adult lifespan from 25 to 92 years, which are representative of the atrophy and neurovascular health characteristics of normal individuals in seven age-bands: 1) 25-34 years, 2) 35-44 years, 3) 45-54 years, 4) 55-64 years, 5) 71-74 years, 6) 75-78 years, and 7) 91-93 years. These atlases can be downloaded from Edinburgh Datashare at https://doi.org/10.7488/ds/1369. The present dataset contains the regions where we can find more incidence of false positives in the segmentation of PVS at each life interval.The brain templates in this dataset, are also part of https://doi.org/10.7488/ds/1369 and https://doi.org/10.7488/ds/3029
Associations between obsessive-compulsive disorder, early life stress, psychological variables and treatment resistance
No full textThis dataset was collected using an online survey between May and July 2022. People with and without obsessive-compulsive disorder (OCD; n = 390) were asked to complete a survey which included the following questionnaires: Obsessive-Compulsive Inventory Revised (OCI-R); Child Abuse and Trauma Scale (CATS; Sanders & Becker-Lausen, 1995); Brief Experiential Avoidance Questionnaire (BEAQ; Gamez et al., 2014); Hospital Anxiety and Depression Scale (HADS; Zigmond & Snaith, 1983); and the "Planning" and "Active Coping" subscales of the Coping Orientation to Problems Experienced (COPE) questionnaire (Carver, 1989). Participants were also asked to provide information on whether they had OCD (and if it was self- or professionally- diagnosed) and what treatment they had received for OCD. Brief demographic information (age, gender, ethnicity, university attendance) was also recorded.Read_me.txt
- This file contains the information below and notes on the content of the dataset
Dataset: Open Data_Associations between obsessive-compulsive disorder, early life stress, psychological variables and treatment resistance.csv
- This csv file contains the data. It contains 390 cases plus a header row.
Data dictionary: Data Dictionary_Associations between obsessive-compulsive disorder, early life stress, psychological variables and treatment resistance.csv
- This csv file contains information on the variables, how they were coded, and which questions they refer to. See "Notes" column for information on reverse coding.
Survey used to collect data: Survey_Associations between obsessive-compulsive disorder, early life stress, psychological variables and treatment resistance.pdf
- This is a pdf copy of the survey that was used to collect the data in the dataset