Strategies for selecting and managing equipment in a light microscopy facility

Light microscopy facilities vary in the number of imaging systems and the scope of technologies they support. Each facility must craft an identity through the selection of equipment and development of staff in order to serve the needs of its local research environment. The process of crafting a light microscopy facility can be compared to curation of an art exhibition: great care should be given to the selection and placement of each object in order to make a coherent statement.


INTRODUCTION
A well-balanced light microscopy facility supports the needs of its local scientific environment by providing a variety of services, from basic imaging support to horizon scanning for new technologies.Crafting a successful facility involves choosing which imaging methods to support and which methods not to support.This requires developing an understanding of the local research environment, institutional expectations, and how these translate into individual and facility assessment criteria.There are many things to consider when choosing the equipment to incorporate into a light microscopy facility.Most systems will come from project or equipment grant funding, which core facilities are increasingly able to influence.The facility head can also decide when, and under what circumstances, to adopt stray microscopes from individual research labs.Finally, retirement of equipment at the appropriate time, through careful assessment of opportunity costs, contributes to facility composition.microscopy facility.These include the grouping of equipment into technology and application areas, the development of staff expertise based on application areas, space considerations, the equipment lifecycle, and other strategic considerations.This article will argue that it is better to support fewer application areas more thoroughly than to offer shallow coverage over a wider range of technologies.

TECHNOLOGY TYPES AND APPLICATION AREAS
It is important to acknowledge that no single microscopy facility can support every available imaging method.In most cases there is an inverse relationship between the number of different methods supported and the effectiveness of support.A facility head must therefore understand the research needs of their local environment and prioritise the selection of equipment and methods to address these needs.In this process it is helpful to think in terms of technology types (such as point-scanning confocal microscopes) and application areas (such as systems suitable for imaging thick samples like spheroids and organoids).Robust support for a given application area usually requires providing different technology types; for example some thick samples can be imaged using a spinning disk whereas others require lightsheet microscopy (see Ref. 1 for a discussion of choosing among similar microscopes).Likewise, robust support for a given technology type generally requires offering more than one system of that type; for example confocal microscopy is better supported by having two point-scanning confocals from different suppliers with different advanced features, or by having a point-scanner and a spinning disk.On both levels, technology type and application area, subtle differences in instrumentation can make all the difference in determining experimental success and it is therefore crucial to provide users with options.However, in a world of limited resources, it will not be possible to support every method equally.It is therefore better to develop real expertise in specific application areas than to offer more limited support across a broader range of methods.'Jack of all trades' sounds nice, but 'master of none' is the kiss of death for a research facility supporting advanced technology.
Grouping of technology types and application areas can also simplify the setting of hourly usage fees by creating price categories. 2 Differences in hourly rates among systems of a similar type (e.g.confocals) can distort equipment use.For example, users may avoid using a more expensive microscope in favour of cheaper options.This can slow the uptake of new equipment or lead to a 'death spiral', whereby underused equipment becomes more expensive to operate as the fixed costs (e.g.service contract) must be recovered over fewer hours of equipment use.This can work against the operation of niche systems, which provide important technological diversity but are not used as heavily as workhorse systems.Grouping systems by technology type or application area into the same price category can simplify this situation by combining more and less heavily used instruments in the same category (although beware some EU funding bodies may not allow this).

STAFF EXPERTISE
What is the right number of staff for a facility?The number of staff per facility has been weakly correlated with the number of users 3,4 and more strongly with the number of systems. 3These studies suggest that most facilities have between 2 and 5 members of staff, with 1 staff member per 30-40 users.The first member of staff typically manages 7 or 8 systems, with another member of staff for every 2.5 systems added to the facility (nSystems = 2.5 × nStaff + 5 systems).The strong correlation between staff and equip-ment supports a facility management model where staff specialise first by technology type and then by application area.
Developing staff expertise according to application areas leads to better service and is consistent with many important career development goals.Staff who specialise in supporting adjacent imaging technologies are better able to advise researchers on the best experimental approach within an application area.It makes sense for a staff member who supports lightsheet microscopy to also support spinning disk microscopy rather than TIRF microscopy, because these methods cover an overlapping application space and may even be used complimentarily in the same research project.Staff who develop expertise in adjacent methods within an application area more fully understand the strengths and weaknesses of the technologies they support and are better able to advise researchers on the best possible approach for their experiments.
From a curiosity and general knowledge perspective staff may wish to have some experience with a wide variety of methods, and within a smaller facility it may be unavoidable for staff to cover a wider range of application areas.However, allowing staff to choose and develop in-depth knowledge of technology types and application areas has many advantages for career development. 5First, specific in-depth knowledge is more likely to lead to substantial research contributions resulting in coauthorship on publications and opportunities for conference presentations.This approach allows staff to build up a CV which demonstrates substantial expertise, rather than more superficial acquaintance with methods.
Second, staff expertise is useful for supporting grant applications, which can be strengthened by listing facility members as co-applicants or named support staff.Funding agencies increasingly encourage this approach because they want assurance that equipment will be managed well and used effectively.The trend towards inclusion of facility staff on grant applications increases the influence of core facilities over their funding and provides important career metrics for facility staff.Finally, developing expertise in an application area creates opportunities for staff to engage in project work beyond basic user training.The generation of high-quality data through advanced imaging methods requires substantial investment of time and effort.Access to facility staff expertise presents options to research labs for how to accomplish specific experimental tasks.It may be more cost effective, and generate better quality data, for facility staff to perform complex low-volume work than to train research lab members to do the work themselves (as is often the case with electron microscopy).Project work can also contribute to a successful feedback loop of acknowledgements, publications, presentations, and funding involvement.

PHYSICAL SPACE
The physical requirements of light microscopy spaces have been thoroughly covered elsewhere in this special issue. 6In terms of equipment selection, different technologies require different spaces and the physical space available will impact the equipment which can be supported.Ideally, systems within a technology or application area should be physically grouped together.It is easier for one person to manage multiple systems if they are near each other, which results in those systems being better supported.Clustering similar systems can therefore be a useful strategy for smaller teams needing to maximise their efficiency.There are also benefits from having users with similar samples and experiments located near each other, where they can potentially interact and exchange knowledge.Thus, there is a synergistic effect of having 3 confocal microscopes in the same room which would be lost if the three instruments were placed in different locations.
Having a central space for the microscopes near to the staff desks makes it easier for staff and users to interact.Equipment located in a dedicated space near staff desks is ultimately better supported, although staff may also benefit from access to a quiet space for some types of work.Conversely, there are often good reasons to place advanced imaging equipment, which inherently requires more support, outwith the facility.For example, the experimental benefits of locating an intravital microscope in the animal unit justify the increased difficulty of support resulting from this location.Biological containment (e.g.BSL-2 or BSL-3) may also require placement of equipment in restricted areas, which introduces many challenges for access and management.Equipment which is distributed around the local institution can be more convenient for researchers to use although it is more difficult to support.Drawing staff out of the central facility to support distributed equipment can lead to beneficial interaction with the wider local environment.One way of better supporting distributed systems is to introduce advanced systems into the facility for their first few years of use and then move them out to a distributed location.This keeps the system near to staff for support during the initial difficult phase of getting a new system up and running, then moves it out to be near its users once it has built up a population of expert users.This approach has the added benefit of freeing up space in the central facility to accommodate incoming new technology.

THE EQUIPMENT LIFECYCLE
When considering the equipment selection process, it is helpful to think about the entire lifecycle of the instrument: demo, pursuit, installation, ramp-up, steady-state, wind-down and de-commissioning (Figure 1, see also Ref. 7).Equipment demos are an essential way to learn about new technologies and a useful way to gauge the scientific interests and needs of the facility user base. 8Relevant new technologies should be tested as and when they become available as part of the facility's horizon-scanning duty, rather than only in conjunction with an intended equipment purchase.A good demo can energise the local research community, motivate funding applications, and provide important preliminary data to support them.Companies may find it easier to justify the cost of a demo based on the availability of existing funding or the size of a facility's user base, but strategies exist to promote general information demos and to support demos at smaller facilities too.For example, a demo may be easier for a company to justify if access can be provided to users from outside the facility, at the city, state, or national level.Local networks, which enable multiple smaller facilities to express interest with one voice, can be especially important for leveraging equipment demos.Finally, a demo involves more than just testing the equipment; it is also a test of a company's application and equipment support teams.
Pursuit encompasses several steps: deciding to pursue a system or technology, pursuit of funding, and procurement.This stage can be difficult to plan exactly and the steps may be interconnected.The first step is deciding what to pursue.The decision to pursue a particular system or technology can be driven by clear internal needs or result through horizon scanning and recognition of opportunities.A consensus must be reached among local PIs, administrators, and the facility head about what system or technology to pursue and how to pursue it: through internal funding, as equipment within a research grant, as a stand-alone equipment grant, etc. Demos can be helpful for reaching a consensus on what to pursue by sensing and developing the necessary level of support among the research labs.A selection decision involves navigation and negotiation of financial realities, subjective preferences, and objective performance criteria.Facility managers rarely make this decision alone and have a critical role to play in promoting opportunities which address the needs of the local environment in a balanced way (not just one loud group) and are likely to succeed as equipment within the facility.Facility managers and staff also play an important role in the pursuit of funding.They may serve as named support staff, co-investigators, lead investigators, or provide letters of support.They are well placed to provide the assurance funding agencies seek that capital investments will be managed effectively.The choice to pursue a technology or a specific piece of equipment will influence the approach to procurement.If the decision is made to pursue a general technology or application The lifecycle of an advanced imaging system.area (e.g. a new confocal microscope or a system to image organoids) then procurement may be approached in the form of an open tender, in which solutions are evaluated and a decision is made during the procurement process.Alternatively, if a specific system with unique features based on patented technology has been evaluated and validated then a single-supplier approach may be justified.Procurement results in a legally binding agreement with the supplier which can provide performance guarantees to ensure the system delivered meets the specifications of the system promised.The details and options around the decision to pursue, acquisition of funding, and procurement will vary according to each facility's local environment.
Installation is a critical part of the lifecycle of any instrument.It begins with a site visit from a company representative to confirm that intended space is acceptable.Facility staff should monitor the installation process carefully.As with the demo, an installation is an opportunity to assess the competence of the engineer installing the equipment as well as the application team providing the initial training.Thorough quality control routines should be performed [9][10][11] by facility staff and any irregularities in the QC results or the installation process should be noted, as these may be indicative of problems to come during normal operation (see Section 6.4).Do not sign off on the installation until you are completely satisfied.During the installation period there may also be opportunities for minor tweaks to the system specification at little to no cost.For advanced systems, training should ideally be divided into 2 or more sessions with time in between for facility staff to begin using the system.
Ramping up is the period of initial use, usually 3-6 months, when new users (and facility staff!) are introduced to the system and begin to understand its features.During the ramp-up phase it is important to enlist appropriate users.Competent users with experience adopting new technologies will be able to recognise the equipment's strengths, identify bugs or other performance issues, which should be addressed prior to full-scale use, and evangelise among the local user base.It is important to manage expectations during the ramp-up phase so that potential new users do not view initial teething pains as fatal flaws.During this time the equipment is unlikely to be performing at full strength due to a combination of technical issues and incomplete understanding of how to use it on the part of users and staff.In a fee-for-use facility, it is therefore ideal if equipment can be used for free or at a discounted rate during the ramp-up phase to avoid friction and dissatisfaction among new and potential users.It is also important to perform quality control tests 9,11,12 during this time to establish a performance baseline for future reference.
Steady-state equipment operation is about maximising value.A web-based solution for equipment management is essential and should be able to capture both equipment booking and use.Booking software should allow selective booking of staff in conjunction with equipment, for example, a user should be able to book a microscope for 4 h with a member of staff for the first 30 min to assist with setting up the experiment.Expectations for equipment use will vary by system type: a system used for overnight imaging will always be booked more than a system which requires a user to sit in front of it.Equipment use will also vary with the academic calendar and the publication cycle of individual projects.A heavily used live-imaging system can be used >100 h/week during intense periods but this might fall to 25 h/week in January or August.Instrument use, and thus capacity, should therefore be estimated on the basis of both average weekly values and peak use.Vital equipment should ideally be covered by a service contract to ensure its productivity and promote longevity.Advanced systems like point-scanning confocals should operate for at least 8-10 years and will often run longer if well maintained.Judicious upgrades can increase the operational lifetime of an imaging system; however, the value of investing in older systems should be carefully assessed.
Wind-down is often an underrated stage of the equipment lifecycle.Even as it approaches end of life, a workhorse system will often still have a dedicated group of users who wish to continue using it for a variety of reasons, including familiarity, continuity of data with previous work, and unique technical features.Users may even protest that there is no reason to stop a running system.However, there are in fact many good reasons to do so!First, there are significant opportunity costs associated with maintaining older equipment (see below).One of the most important of these is space: at some point old equipment needs to be retired to make space for new equipment.Second, service contracts cost money.The 'death spiral' mentioned above can be mitigated by reducing the level of coverage; however, when the cost of the contract is no longer covered by the level of equipment use, it is time to begin winding the system down.If a research lab insists that an aging system cannot possibly be retired, it may be worth asking if they would like to have it in their lab and pay for the service contract.Finally, a system will eventually reach end of life and no longer be supported by the manufacturer.At this point the risk increases that it will break and be unrepairable, potentially leaving researchers in the lurch.It is preferable to manage the migration of users away from an aging system onto new equipment in a controlled fashion, rather than simply waiting until the old system dies.Once the old system dies, there is no opportunity to cross-check samples if confounding experimental parameters, such as changing serum or other reagents, give rise to unexpected results.The first step of the wind-down process is to stop training new users.This should be done by setting and announcing a date, which will help existing users to appreciate the reality of a system reaching end of life.The next step is to remove the system from the booking system.Again, this should be done after setting and announcing a date, ideally with the unannounced intention that the system will remain in place for a month or so after it is no longer bookable (just in case).
Decommissioning and removal are often painful steps.It can be difficult to accept that a system bought for hundreds of thousands of dollars, and producing useful data just yesterday, is now worth nothing.The possibility of selling or donating old equipment may be attractive; however, in most cases, the factors which led to decommissioning in one facility (low use, outdated features, poor reliability, manufacturer's end of life, etc.) will be equally prohibitive if the equipment is moved to another facility.Shipping old equipment to resource-constrained research environments (e.g. in the global south) can backfire and should be approached with caution. 13If the plan is to sell old equipment beware, this may take time: the best approach is to find a company who will first remove it and then reimburse the facility later if they manage to sell it.Another option is to find someone with the same system who intends to continue using it and donate your system for spare parts (although, frustratingly, it can be administratively more difficult to donate old equipment than to throw it away).In most cases some useful components can be stripped from a decommissioned system prior to disposal, including objectives, filters, and electronic components such as shutters, stages and lasers.Making labs and home-builders are often grateful for specific components.Note that suppliers within the EU may be required to dispose of electronic equipment they provided.

6
OTHER STRATEGIC CONSIDERATIONS

Workhorses, racehorses, and white elephants
It can be useful to think about the systems in a microscopy facility on a spectrum between workhorses and racehorses.Workhorses typically offer more robust performance due to the use of basic technology whereas racehorses offer more advanced technology and temperamental performance.In simple terms, one can think of the role of a facility manager as balancing the equipment selection of the entire facility on this scale: while the facility as a whole must be seen as a workhorse for the local institution, it is the racehorses which define how a facility is judged with respect to the cutting edge.A few systems manage to offer both robust performance and cutting-edge technology.Unfortunately, even such an ideal system can still become a white elephant in the wrong environment.The conventional definition of a white elephant is something expensive and useless.However, useless can either refer to technology which is poorly designed and implemented or to good technology applied to the wrong application area.An important filter for preventing the acquisition of white elephants is to insist that no microscope should be bought without the evidence of convincing demo data using actual user samples.

Beware of Swiss army knives
In a research grant-driven funding environment there is a tendency for research group leaders to specify microscopes with all the features available to make the most of rare opportunities for large equipment purchases.Equipment suppliers are only too happy to support this practice by specifying and selling 'fully loaded' systems.This approach may make sense for an individual research lab but rarely makes sense for a microscopy facility.For example, a research lab of 10 people may be quite happy to have a dual-use microscope capable of performing both confocal and lightsheet microscopy because the usage demands on the system are modest.However, it is important to remember that '2 for the price of 1' also means '50% of each' when it comes to equipment capacity: a dual-use system provides only 50% capacity for each method.In a microscopy facility catering to larger numbers of users, it is a better strategy to pursue funding for microscopes dedicated to each individual method.Systems purporting to offer access to two different imaging modalities in one (e.g.confocal superresolution combined with multiphoton excitation and nondescanned detection) should be approached with caution.Such systems have an operational risk that if one component breaks it can effectively knock both methods out of operation, affecting two groups of users at once.This risk is compounded by increased likelihood of a more complex system breaking down in the first place.Finally, a multifunctional system will be more expensive and therefore more difficult to clone if more capacity is needed.

Single versus multisupplier approaches
There are advantages and disadvantages to preferentially selecting equipment from a single supplier 14 or running a branded microscopy facility. 15A selective strategic partnership can promote early access to new technology and/or increased service support from the partner company.It is easier for staff to support equipment from the same supplier (e.g.only one set of features and bugs to remember, greater interoperability of components, etc.), which can make this an attractive option for facilities with fewer staff.It can also simplify user training if most of the systems in a facility run on the same software.Finally, preferential interaction with a single supplier means that attention can be focussed on developing relationships with one set of sales, application and service support staff.In contrast there are many disadvantages to using a single-supplier approach, starting with limited access to technology.Access to cutting-edge technology is availed by using the full range of suppliers in the market.Different companies develop different products based on their corporate culture and strategies, and the corresponding intellectual property they acquire.Even within the same technology type (e.g.multiphoton microscopy), companies operate on different product development cycles.Technology offered by one company today may be superseded by tech from another company tomorrow.Another advantage of selecting equipment from the full range of available suppliers is competition; suppliers must compete in terms of both price and the provision of service.It is a powerful leverage to show a rep from company X a system from company Y and ask, 'Why can't your system do that?'A good awareness of the equipment offered by different suppliers and specialisation in specific application areas can also increase the attractiveness of a facility for demos and early access to new technology.User preference is another practical reason to diversify: some users will come preconditioned with personal, experimental or data-continuity preferences for equipment from a specific supplier and it often makes life easier to accommodate these rather than insist on users changing their approach.

Quality control
Regular quality control of advanced equipment is essential because inexperienced users may not recognise when its performance has been compromised.In the absence of quality control, 'rigor and reproducibility are limited, the reliability of quantitative analysis is severely impacted and the confidence in published data becomes eroded'. 10Different quality control regimes may be needed under different circumstances.Steady-state quality control of a confocal microscope must aim to test a reasonable number of instrument parameters within a practical timescale of 1-2 h.In contrast, quality control assessments performed during equipment installation and ramp up should be exhaustive benchmarks.In this case it is useful not only to measure the physical resolution of an instrument to determine its accuracy, but also to perform time series measurements over many hours using multiple XY and Z positions to control for precision.The literature in this area is currently evolving rapidly; however, a good starting point is Ref. 11 and the Quarep-Limi website.

Opportunity costs
A facility manager has an obligation to maximise the value of the equipment under their care.This would seem to suggest extending the use of a system for as long as possible.However, the retention of underused equipment incurs significant opportunity costs for the facility and detracts from its efficient operation in several ways.One obvious problem is that the service contract on any system becomes progressively more expensive (per hour) to recover the less the system is used.Underused equipment can also become more difficult to support as staff become less familiar with it and must remind themselves of features and bugs on each (rare) occasion that they use it.Third, underused equipment takes up physical space and can therefore give the impression that the facility does not need new equipment.There may sometimes be value in using old equipment to hold on to floor space for the facility (i.e.space wars), but there are costs associated with this too: a system which is only used a little is potentially blocking space for a system which could be used a lot.Finally, underused equipment takes up mental space.It occupies staff time in disproportion to its use, potentially slowing their engagement with newer technologies.It may also inhibit people from pursuing new funding opportunities based on the perception that facility staff are at capacity and unable to support more equipment.

CONCLUSIONS
A light microscopy facility should be more than a random collection of microscopes.Funding is the major bottleneck limiting the acquisition of new equipment; however, a facility head can significantly influence the selection of equipment allowed into and removed from their facility over the long term.This long-term influence requires a consistent strategy, in order to craft an identity for the facility, promote staff development and effectively serve the local research community.