Workshop 2022
The Annual Meeting of ORSO will take place online (Zoom) on 7th-10th June 2021. The purpose of the meeting is to discuss topics of interest to the ORSO community. The program will consist of scientific and technical talks together with discussion forums and/or activity based sessions for each of the working groups Since the agenda is scheduled to try to cover all time zones, we acknowledge that it may be difficult to attend all these sessions. We will therefore record as much of the meeting as possible and post the videos on the ORSO youtube channel. We will be running an electronic vote on at least 1 issue during the meeting. The topics of the votes will be notified to registants prior to the start of the workshop and presented in the first session before votes accepted electronically during the week.
For those interested a report from last year’s meeting is here
Registration is now open
Agenda:
The agenda is not yet finalized but the confirmed details are below and here using the european time zone.
Tuesday 7th June: Plenary Session (Chair: Tom Arnold)
09:00-11:30 (UK), 10:00-12:30 (Europe), 05:00-07:30 (US East coast), 18:00-20:30 (Australia East coast):
- General introduction to the meeting and how it will work
- Wiebke Lohstroh: Introduction to the DAPHNE project
- The AGM. Explanation of ORSO organisation, voting methodology and election of 4 chairs (1 from each working group)
- A presentation of ORSOPY. We will be voting on the release of this as the first major output from ORSO – please review the osopy specs here
- RECORDING OF THE PESENTATION OF ORSOPY
- SLIDES
Recordings of the talks will be uploaded and made available to all to ensure access from all time zones.
Tuesday 7th June: Collection of Teaching Materials (Chair: Andrew McCluskey)
17:00-20:00 (UK), 18:00-21:00 (Europe), 12:00-15:00 (US East coast), 02:00-05:00 (Australia East coast):
The aim of this session is to build on the existing work at https://www.reflectometry.org/learn/, a short introduction to reflectometry fitting.
In this session, we will discuss what material would be useful for the webpage. There will bean instruction on how to add material to the page and then we will work in small groups to develop new material for the webpages.
The specific areas of focus will depend on the attendees but some potential areas of work include:
- Tools for analysing data
- Expanding Bayesian analysis section
- Updating the X-ray reference material
- Proof reading
The aim in this session is to update the webpage live in a “hackathon”. Please come prepared by reading through the current material to identigfy information gaps or bad links and potentially getting a Github account.
You do not need to be confident with Github to participate, what we need is expertise in reflectometry to ensure that the material generated is of high quality and appropriate for the whole ORSO community.
Wednesday 8th June: ORSO Webpage Workshop (Chair: Tom Arnold)
09:00-11:30 (UK), 10:00-12:30 (Europe), 05:00-07:30 (US East coast), 18:00-20:30 (Australia East coast):
The aim of this session is to review the ORSO webpages and update them live in the session according to input.
We will begin with a short talk explaining how you can edit the pages. Then we will ensure that all those present have permission to do so.
Following a short discussion to identify topics, we will work in small groups to implement changes, collect information and actively modify pages.
The specific areas of focus will depend on the attendees but some potential areas of work include:
- Bibliography
- Beamline lists and publications
- Site navigation
Wednesday 8th June: Priors and posteriors paper (Chair: Andrew McCluskey)
14:00-16:00 (UK), 15:00-17:00 (Europe), 09:00-11:00 (US East coast), 22:00-00:00 (Australia East coast):
Introduction talk “Interperting reflectrometry data with Bayesian analysis” (Andrew McCluskey)
More and more Bayesian analysis is being used to interpret reflectometry data. However, how this information is reported is inconsistent potentially limiting the replicability and reproduciblity of the results. To improve this, within the Reproduciblity Working Group, we have started to develop a short publication defining some best practice for reporting results and inputs to Bayesian analysis.
In this session, we first present and then seek feedback on the current draft of the paper and will aim to improve the work based on this feedback. Note: the expectation for attending this session is that you have read the draft in advance and are able to offer feedback, it is not expected that you are an expert user of Bayesian analysis (as you might be the exact audience that we are targeting in that case). Attendence of this session will lead to inclusion as a co-author of the paper (if desired).
Thursday 9th June: Plenary Session (Chair: Jos Cooper)
09:00-10:00 (UK), 10:00-11:00 (Europe), 05:00-06:00 (US East coast), 18:00-19:00 (Australia East coast):
Thursday 9th June: A Simple model language (Chair: Jochen Stahn)
10:00-11:00 (UK), 11:00-12:00 (Europe), 06:00-07:00 (US East coast), 19:00-20:00 (Australia East coast):
- One of the early aims of the Data Analysis Workig Groups was to create a common vocabulary for the definitionn of reflectometry models.
- This session will demo the model languages that have started to be developed and will conntinue with a discussion
- further details to be confirmed
Thursday 9th June: The ORSO Scattering Length Density Database (Chair: Artur Glavic)
14:00-16:00 (UK), 15:00-17:00 (Europe), 09:00-11:00 (US East coast), 23:00-01:00 (Australia East coast): (please note the error in the Australian time zone for this session has now been accepted)
The aim of this session is to launch the ORSO Scattering Length Density Database. A database exists but we now need to add the necessary data to make it useful to the community. The intention is that this database is maintained by the people that use it, so we need your input to make it work.
- Presentation of the status of the SLD database (Artur Glavic)
- Information gathering excercise and release of the database
- Live testing followed by questions, answers and discussion
- further details to be confirmed
You are welcome to send feedback before or after the session
Thursday 9th June: Plenary Session (Chair: Max Skoda)
16:00-17:00 (UK), 17:00-18:00 (Europe), 11:00-12:00 (US East coast), 01:00-02:00 (Australia East coast): (please note the error in the Australian time zone for this session has now been accepted)
Abstract: For an arbitrary neutron reflectometry (NR) experiment, information regarding physical measurables of interest is not uniformly distributed in momentum transfer (Q) space. Thus, some regions of Q space are more advantageous for measurement than others, providing an opportunity for optimization of data collection rates. The optimal conditions, however, are not in general known before the data are collected and analyzed. Here, I demonstrate an active learning algorithm for driving autonomous neutron reflectometry experiments. Within the context of a user-supplied model, the algorithm uses the data that have already been collected to predict the measurement condition that will maximize the information gain rate for the parameters of interest. Data are collected at the optimal measurement condition, and the process repeats until the experiment is terminated or the model parameters are sufficiently constrained. Application of this technique to model systems shows that autonomous data collection rates exceed those of conventional, step-by-step data collection schemes by several times.
Abstract: In this talk, the advantages of a 1D detector over a point detector for the recording of reflectivity curves at a standard laboratory x-ray source will be discussed. The use of 1D detectors allows one to verify the quality of the recording, correct for shifts due to sample misalignments and gain additional information as e.g. about the diffuse/off-specular scattering. A Python program is developed to automatize the data reduction as well as providing a first hands-on evaluation. It is thought as a potential standard approach for making the recording of reflectivity curves at such lab sources (and others) more reliable, giving an faster and easier access also to non-experienced users of this technique.
Friday 10th June: Validation datasets (Chair: Andrew Nelson)
08:00-09:00 (UK), 09:00-10:00 (Europe), 04:00-06:00 (US East coast), 17:00-18:00 (Australia East coast):
- Introduction – context of existing validation of calculation kernels (with link to https://github.com/reflectivity/analysis/tree/master/validation)
(20 mins)
- Brian Maranville: Recent work on development of validation of calculation kernels for Polarised Neutron Reflectometry
- Discussion of what kinds of systems would be good for PNR validation set
- Discussion of validation datasets for targetting XRR (e.g. significant absorption), especially RSoXRR
- Implementing of PNR validation for different packages
- Transfer of validation datasets to ORSO format
(20 mins)
- Standard Samples
- David Cortie: A standard sample for PNR
- Christy Kinane: Validation samples used at ISIS
- Outlook - discussion on proposals for other areas that could benefit from a similar approach. e.g. anomalous X-ray reflectivity
(20 mins)
- Andrew Caruana: Systematics in laboratory XRR data: Update and progress on data reduction and methodology
Friday 10th June: General description for resolution functions (Chair: Andrew Nelson)
09:00-10:00 (UK), 10:00-11:00 (Europe), 05:00-06:00 (US East coast), 18:00-19:00 (Australia East coast):
The collective action for nomadic small angle scatterers, CanSAS (https://www.cansas.org/) will present their work on resolution functions:
The numerical convolutions of the appropriate functions for fixed wavelength SANS gives improved Q resolution functions that are slightly flatter in the peak and broader in the tails than the classic Mildner & Carpenter approach. However, in time-of-flight SANS the resolution curves from all the wavelengths contributing to a particular Q value are averaged together, weighted by the number of neutrons detected at each wavelength. Since the resolution width is inversely proportional to wavelength this often gives a curve that is noticeably sharper in the peak and broader in the tails than a simple Gaussian.
In either type of SANS measurement a more detailed description of Q resolution smearing within fitting programs such as SasView will need generic resolution functions of arbitrary shape.
The generic resolution function was subject of discussion at CanSAS workshop in March 2021 and efforts have been for devising data format and using arbitrary function in SasView. The summary of these activities will be provided in the presentation
- presentation of the work being done by Cansas (Wojciech Potrzebowski)
- discussion of relevance to ORSO and the potential for collaboration
Friday 10th June: Concluding Plenary Session (Chair: Bridget Murphy)
15:00-17:00 (UK), 16:00-18:00 (Europe), 10:00-12:00 (US East coast), 00:00-02:00 (Australia East coast):
In the final session of the workshop we will summarise the working group sessions and announce the results of the electronic voting.
- Summary of the weeks sessions and next steps
- The AGM voting results: The motion was passed with 19 votes in favour and 1 against.
- Alessandra Luchini: “Developing advanced models of biological membranes to investigate protein-lipid interactions”