Meetings Page/Facility Manager Meeting 2010

From BioImagingUKWiki

Jump to: navigation, search

This year's Microscopy Facility Manager Meeting was held in London, at Imperial College on 7-8 January 2010.

Contents

Summary and outcome

You can now download a full summary of the meeting as a PDF file. This also contains the result of the little survey of participants on content and format, and summaries of the various break-out sessions.

The break-out sessions will feed directly into the BioimagingUK working groups, for the latest updates see the respective subpages of BioimagingUK

Training

CHAIR: Yan Gu, MRC Mill Hill, London

Issues raised:

  • training courses
  • company involvement
  • funding for training
  • training of staff and supply of new staff
  • modular training could even out different background (physicists, biologists) - basic modules (e.g. optical physics, sample preparation, …) could bring everybody up to the same level as an entry point for advanced courses; after completing a certain amount of modules, a certificate could be created
  • central website - Oxford already had twice as many registrations as spaces
  • funding for courses - limited to young people, not facility people?

High-throughput microscopy

CHAIR: Nick Barry, Laboratory of Molecular Biology, Cambridge

  • The difference between low throughput (tens of samples with data collection on standard microscopes) and medium throughput (hundreds of samples, multiwell plate format, some degree of automation) is not so big: people already do repetitive experiments that could be automated.
  • High-end plate reading instrumentation is entering facilities (e.g. P.March, Manchester)or a closely affiliated facility (e.g. T Self, Nottingham). With the provision of some robotic sample handling and up to medium scale sample numbers, the user can take care of reagents and sample prep. At this level it can function within a well resourced microscopy facility.
  • Assay development for HT generally is done via this ‘hands-on’ route.
  • High-throughput is different, with thousands to tens-of-thousands of samples. The whole experimental pipeline needs to be highly automated, with a lot of support/ancillary equipment and in house expertise.
  • Setup costs are high, but numbers of data points are huge as well
  • Expert support staff is needed, other than the person performing the experiment: biologists, bioinformaticians, statisticians, I.T. people
  • Data analysis and bioinformatics are essential,
  • Some public domain software is around, what’s missing is TRAINING!!
  • Some groups report shortfalls with existing data analysis software - method and software development is needed
  • Large HTS facilities could save money by buying libraries/reagents in bulk, and could also more easily set up the infrastructure to store them safely. Ultimately a large screen is still expensive
  • From the perspective of bringing automated microscopy into wider and more routine use, hardware and software developed for HTS has a lot to offer.


See also working group Higher speed/content/throughput functional imaging

In Vivo Imaging

CHAIR: Paul Thomas, Univ. East Anglia

Also present at the break-out session: Ben Atkinson, Intelligent Imaging; Grant Calder, John Innes Centre; Dave Clarke, Rutherford Appleton Laboratory, STFC; Paul Cormack, Hamamatsu Photonics UK Ltd.; Ana-Maria Calcagno-Pizarelli, Imperial College London; Roy Edward, Biostatus Ltd; Felicity Gavins, Imperial College London; Trudi Gillespie, Univ. Edinburgh; Colin Gray, Univ. Sheffield; Peter Humphreys, WT Centre for Stem Cell Research Cambridge; Peter Jordan, CRUK London; Emma King, Univ. Dundee; James McKechnie, Media Cybernetics UK; Luis Pizarro, Imperial College London; Kees Straatman, Univ. Leicester; Rolly Wiegand, Univ. Edinburgh.


A small group (listed above) met to discuss the use and development of tools for more physiological (in vivo) imaging. The break-out session was followed by a round-table discussion involving all the meeting attendees. Below are the major points that arose out of both the break-out session and the subsequent summary session:

General in vivo imaging

  • The need for a central facility providing access to all UK scientists was a major discussion point. Overall it was felt that there are already a large number of facilities across the UK which provided sufficient access to external scientists; and thus, a central facility didn’t make much sense.
  • Nevertheless, it was felt that the availability and accessibility of in vivo imaging facilities might be better advertised; perhaps, through the UK LM facility website.
  • Furthermore, it was also thought that the UK LM facility website could indicate which facility managers were willing to provide help and advice (applies to other areas of expertise as well as in vivo).
  • Even considering the fact that the majority didn’t support the idea of a central facility, it was acknowledged that one disadvantage of local in vivo imaging is that in vivo and standard microscopy can often exclude each other; thus, causing duplication of equipment.
  • It was also felt that centres of excellence dedicated to specialized forms of in vivo imaging (similar to the centres established by the Scottish Universities Life Sciences Alliance; SULSA) might be a good idea. The major advantage of the SULSA model being the equal charges and accessibility applied to all investigators; thus, providing facilities for those scientists at institutes that have no in vivo capability.
  • Another major point of contention was the problem of Home Office licensing. There was overall dissatisfaction with the inconsistency of different Home Office inspectors in applying the same rules in different ways. Likewise, there was a feeling that licences were overly restrictive; the fact that a licence is linked both to an experiment and a location was a big problem. If the Home Office would change it so an experiment could be done in any licensed premises, that would help.

Cat-3-specific points

  • Unlike general in vivo imaging, it was felt that a central facility for Cat 3 work is needed; perhaps, at site that has most safety procedures already in place – e.g., Pirbright (animals) and Rothamsted (plants).
  • However, it was pointed out that a Cat 3 licence is always specific for an organism, two Cat 3 experiments can’t necessarily share the same space, leading to duplication of equipment.


See also working group In Vivo Imaging

Funding and Sustainability

CHAIR: Lucy Collinson, CRUK London

  • Direct funding for open access facilities, incl. service contracts and support staff, rather than via research groups, would be more efficient (better usage of equipment) and induce a change of mentality(no protectiveness and secrecy).
  • Imaging inventory:
    • An inventory of facilities across UK would be good, as a source of information for all scientists on what equipment and know-how is generally available, and for universities and funding agencies to dedicate funding.
    • This should include a wishlist for equipment, as a reference for future planning.
    • The inventory could build on the list of UK light microscopy facilities currently maintained by the imaging facility at the University of York
  • The balance "small - medium - national facilities" needs to be discussed, to avoid that large central facilities don’t take money away from established local ones.
  • Specialised equipment and pre-commercial novel technology:
    • Dedicated funding for specialised equipment and pre-commercial novel technology is needed, to promote development.
    • How can companies be involved?
    • Would a mixed Open Licence / Commercial Licence model for technology be valuable, as is already established for software?
    • Could companies save money by sharing development costs with funding bodies, which in exchange get the final product cheaper?
  • Currently, equipment funding is channelled via project grants, but the most efficient use of equipment is outside / across single projects - different metrics to review funding for facilities are needed.
  • Bulk service contracts between research councils and companies would make service contracts cheaper for facilities and avoid wasting money on equipment soon to lay idle, but the number of service contracts would most likely be higher and more predictable, making it attractive for companies.


See also working group Sustainability

Software and Data

CHAIR: Paul Thomas, Univ. East Anglia

Also present at the break-out session: Ben Atkinson, Intelligent Imaging; Grant Calder, John Innes Centre; Colin Gray, Univ. Sheffield; Ki Hng, MRC Clinical Sciences Centre; Deliza Ibanez-Garcia, Bitplane AG; James McKechnie, Media Cybernetics UK; Anton Page, Univ. Southampton; Luis Pizarro, Imperial College London; Tim Self, Univ. Nottingham; Alex Sossick, Univ. Cambridge; Kees Straatman, Univ. Leicester; Simon Walker, Babraham Institute.


The small group (listed above) met to discuss the use and development of software tools for data management, archiving and analysis. The break-out session was followed by a round-table discussion involving all the meeting attendees. Below are the major points that arose out of both the break-out session and the subsequent summary session:

  • The Facility managers overwhelmingly supported moving to a single, standardized file format (e.g., OME-TIFF). This idea was also tentatively supported by representatives of companies developing image-analysis software. Nevertheless, it was thought that a stumbling block would be the acceptance of this idea by image-acquisition companies.
  • One objection raised by imaging software developers was that the present OME-TIFF format does not encompass all the necessary meta-data.
  • There was general support for an on-line, centralised database for published images that includes all original images generated in support of a publication (similar to the JCB DataViewer, but global).
  • It was thought that such a database would be useful for validation of data by peers, and would also enable software developers to check new algorithms on “standard” datasets.
  • There was less enthusiasm demonstrated for the adoption of the OME database (Omero). In the past, some managers had tried Omero, but had failed to get it to work. These managers were disinclined to try again with the newer version despite assurances from managers with newer installations that it was now much improved.
  • Those managers using Omero argued that it enabled better organisation of large amounts of data, and simpler access to archived images.
  • The majority of managers were not users of Omero – perhaps, the biggest impetus to uptake will come if the centralised database is implemented forcing imagers to seriously consider archiving their data in an easily searchable and accessible repository.
  • Additional impetus for implementation of Omero (or something similar) may also come from funders that begin to seriously apply their rules regarding data archiving and access.
  • Those who use Omero (as well as those who had tried) felt that uptake of Omero might be greater if it was simpler to install and had more image analysis tools.
  • One of the problems identified with Omero is the data duplication problem where data is kept in its original format and as OME-TIFF in the Omero database, increasing the data storage required for data and backups.
  • There was much discussion of commercial alternatives to Omero (e.g., Imagik). The general feeling was that there could be problems of data retrieval if the company were to collapse, or if users were no longer able to afford the service. It was thought that this was less of a problem with open-source software such as Omero.
  • A separate issue was how, and to what degree, can experimental (as opposed to image) meta-data be standardised and saved along with the images (e.g., sample preparation, organism, tissue, etc.)
  • Many managers thought that more workshops aimed at software training would be useful; especially, for open-source software (ImageJ, Omero, µManager). Although, it was pointed out that RMS runs image processing workshops every year (June) – and these include ImageJ.
  • Idea for next year’s meeting: Omero workshop.


See also working group Software Tools & Data Management

High-Resolution Microscopy

CHAIR: Rolly Wiegand, Univ. Edinburgh

Over the last 3 years, a range of novel microscopic techniques have become available that circumnavigate the limitations as defined by Abbe’s law of diffraction-limited resolution of optical microscopy, thus providing ‘super-resolution’. These new approaches were discussed and here a short summary of the methods is given:

Structured illumination microscopy (SIM)(OMX / Zeiss PAL-M):

  • ~2x increased XYZ resolution (~100 nm lateral, 250 nm axial)
  • too slow for live cell imaging in SIM mode
  • limited penetration depth (< 20 m)
  • strong signal necessary (photobleaching might be a problem)
  • costs are high, therefore only affordable for large facilities/institutes (at present Oxford and SULSA/Dundee)
  • no multi-colour image acquisition in high resolution mode (grid wavelength-dependent, can’t be easily realigned)
  • Zeiss PAL-M has added functionality of PALM and costs might be slightly lower

4Pi:

  • Leica have discontinued the system and vast operational problems strongly suggest that this technology is not feasible for practical use

Stimulated depletion microscopy (STED):

  • 3x increased resolution in XY (~75 nm), no improvement in Z, yet
  • Leica have multiphoton and continuous wave version available
  • purely optical, so no computing involved, but fluorophores need to tolerate high (depletion) laser power and many excitation / depletion cycles
  • modular system allows use of standard scanning systems and thus combined functionality and lower costs
  • acquisition speed as for SIM relatively slow
  • so far only single channel acquisition, but a two-channel solution is under development for the MP-system
  • CW-system allows the use of standard fluorophores and provides less problems with photobleaching
  • easy to operate by standard user

Statistical methods (PALM, STORM, …):

  • 3-4x increased resolution in XY, no improvement in Z in basic mode, therefore most applications combine it with TIRFM so far
  • modified methods (e.g. astigmatic optics with 3D STORM, de-focussing or double-lens systems (iPALM)) allow ~3x increased resolution in Z as well
  • relatively easy and cheap to implement on a standard high-end widefield system
  • needs either switchable/activatable fluorophores or special reducing conditions for ground-state depletion of standard fluorophores (so far limited to fixed samples)
  • requires very stable, high precision stage and environmental conditions
  • high computational demand for data processing
  • first publication of two-channel image acquisition
  • can be combined with standard widefield image acquisition/deconvolution to add information from a different spectral channel at ‘normal’ resolution

General topics

  • a discussion platform and workshop would be desirable
  • a workshop has been proposed and initiated:
    • invited speakers could be :
      • PALM / STORM: Sam Hess, Eric Betzig, Jennifer Lippincott-Schwartz
      • STED: Stefan Hell or colleages
      • switchable proteins: Uli Nienhaus, S Jakobs
      • application of the techniques above: Anne King and/or Ian Dobbie (OMX), Christina Flors (TIRF-PALM, ground-state depletion microscopy), ….
      • microscope companies at the meeting (Zeiss and Leica) are happy to provide their demo equipment (Zeiss: PAL-M, Leica: STED, Olympus: Cell R system capable of TIRF-PALM , other widefield system suppliers for PALM?); this would allow to compare all techniques with the same sample side by side
      • location to be decided, if the OMX system is to be used, it would have to be in Oxford or Dundee, because it can’t be transported
    • a good date would be beginning of July, after the equipment is used at Microscience
  • The feasibility of a central research and development facility for super-resolution microscopy was discussed (desirable but presumably impossible to fund and administer).
  • Due to the increasing demand in computing power, data management and storage capacity, the improvement of local IT infrastructure and collaboration with computing experts seems to be the way forward for central imaging facilities (see also break-away session ‘Software and Data’).


See also working group Higher Spatial Resolution

Careers

CHAIR: Peter March, University of Manchester

  • most people ‘ended up’ as facility manager
  • nobody had any training on actual facility management, as of management, legal implications, health and safety
  • acknowledgments and publications are a constant problem
  • big problem, and urgently needed: lack of job profile for facility manager within the academic structure, with career progression, recognition, involvement in teaching, development, …
  • this has to be brought up with colleges, universities, funding bodies
  • a new job profile to be pushed for
  • how can facility managers (and facilities) be reviewed, if not by publications (e.g. user surveys?)


See also working group Careers

Programme

=> volunteers to chair break-out sessions needed!

=> feedback is still very welcome! (m.spitaler@imperial.ac.uk)

Thursday 7 January 2010

  • 12.00-13.00 Arrival, sandwich lunch and registration for break-out sessions
  • 13.00-13.15 Welcome, latest updates on programme
  • 13.15-13.30 Introduction of participants
  • 13.30-14.15 Short reports & discussions
    • Paul Thomas, UEA: Multiphoton microscopes
    • Peter Humphreys: STJ detectors
    • ad hoc topics
  • 14.15-14.45 Technical Report: James Francis (MediaCybernetics): "How to quantitatively test and compare cameras" (the practical demonstration announced previously has been snowed in and is cancelled)
  • 14.45-15.15 coffee
  • 15.15-15.30 Update BioimagingUK / Eurobioimaging
  • 15.30-15.45 Introduction Break-out Sessions I (chaired by...):
    • Training (Yan Gu)
    • High-throughput microscopy (Nick Barry)
    • In vivo imaging (Peter O’Toole)
    • Funding & sustainability (Lucy Collinson)
  • 15.45-17.00 Break-out sessions I
  • 17.00-18.00 Summary of break-out sessions by session leaders, discussion
  • 18.00-19.00 hotel check-in
  • 19.00-late dinner

Friday 8 January 2010

  • 9.30-9.45 Introduction to Break-out Sessions II (chaired by...):
    • Software & Data (Paul Thomas)
    • General facility access (?)
    • Access to specialised, pre-commercial equipment (?)
    • High-resolution microscopy (Rolly Wiegand)
    • Careers (Peter March)
  • 9.45-11.00 Break-out sessions II
  • 11.00-11.30 coffee
  • 11.30-12.30 Summaries of break-out sessions + final discussions
  • 12.30-13.30 lunch and option for FILM facility viewing


=> Doodle Poll: people's interests in topics

Date, time and location

The 'Doodle Poll - meeting dates' showed that many people can only do the first week of January, so the definitive date will now be Thursday 7 (lunch time) to Friday 8 January 2010 (lunch time).

The meeting will be held in the South Kensington campus of Imperial College London, 5th floor of the Sherfield Building / room SALC 5 (see Fac_Man_Meeting_2010_directions.pdf or GoogleMap for details). Walk into the campus from Exhibition Road, just around the corner from the Science Museum, and walk around the Queen's Lawn (with the big tower) in the middle of the campus.

This is ~30 min by public transport from Euston and Kings Cross railway stations, 10 min from Victoria station, or a 25 min walk across Hyde Park from Paddington (slightly longer on the Circle Line). For time tables and other stations, see Transport for London (TfL) Journey Planner.

Accomodation in walking distance from the meeting location can be booked at discounted rates (starting from £69) through the Imperial College Hotel Accomodation website. Many affordable hotels can also be found on South Kensington Hotels, many of them actually cheaper than through the Imperial College website and also in walking distance, but you need to check distance (e.g. TfL) and quality (e.g. Tripadvisor) yourself.

Travel and weather update

Update Thu 7 Jan 8:33:

Most trains across the country seem to run with only minor delays (under an hour), so the meeting will go ahead, maybe with a little extended lunch to give people more time to arrive. For travel updates see Transport for London, TravelDirect (train) and BAA (airport). The weather forecast for today and tomorrow is also pretty good (Metoffice and MetCheck).


Registration

Online registration is now closed, if you still want to attend please get in contact with m.spitaler@imperial.ac.uk

A list of participants is available on the registration page

Personal tools