Plant Sample Prep for Nuclei Analysis

We don't have too many Botanists coming through the flow lab doors, but occasionally they'll bring in some electric green looking slurry of stuff and want to analyze DNA content or Mitochondrial Membrane Potential. It's really no different setting these samples up on the instrument than any other animal/human sample. But, when they start to ask me questions on how to optimize their sample prep, I'm of little help. From what I understand, the process is much like dissociating any tissue. You use some mechanical (slicing and dicing with a razor) and enzymatic processes to try and isolate the cells from the leaves. Beyond that basic understanding of what's going on, I have little to offer, which is why I appreciate this troubleshooting guide so much. A fella by the name of Paul Kron, from the Department of Integrative Biology at the University of Guelph (Guelph, Ontario, Canada) sent me his top 10 troubleshooting tips for achieving good quality histograms when analyzing DNA content from plant material. There's probably some things in here you'll think are somewhat elementary, but isn't it always those elementary steps that are lacking when things aren't going quite right? A short, easy-to-read guide like this will ensure you've thought about some of the minutia involved in optimizing your DNA profiles from plant materials. A link to the top 10 can be found here: http://www.uoguelph.ca/~pkron/PaulKron/Top_10_FCM_Tips.html.

Do you Kaluza?

Seems like there's quite a bit of news coming out of Miami these past few months, and that is a good thing. Beckman Coulter (coulterflow.com) has released it's offline analysis tool, Kaluza, whose major feature seems to be speed. I've been playing around with Kaluza for about a month now and so I can share a few thoughts. You're probably well aware of the fact that we at the University of Chicago are a FlowJo shop. Nearly all of our users analyze with FlowJo (others use acquisition programs like CellQuest or FACSDiVa to do analysis). We've looked into other packages in the past (namely FCS Express and VenturiOne) but none were as exhaustive in capabilities as FlowJo, so we felt we were getting the most bang for buck in that case. What this basically means is that I pretty much look at all analysis software through a "FlowJo-colored lens." My impressions therefore, are fairly skewed towards that bias. With that being said, here's my first impressions on the windows-only (so sad), late beta version of Kaluza.

First of all, it is aesthetically pleasing to look at and interact with. They successfully pulled off the minimalist/high-tech look and feel of the software. Getting started with the software is easy enough. You load FCS files, apply analysis plots/regions/stats and that's it. I've loaded small (10K cells, 2 or 3 colors) files as well as quite large (5M cells, 10 colors) files. They initially load about as fast as you'd expect, within a few seconds up to 30 seconds, but once they're loaded, you can make adjustments, move gates around, recalculate stats very quickly. Moving around within the software is very snappy. This is a welcomed change of pace from FlowJo, which feels like an eternity when manipulating things like compensation on many large files at once. The second thing about Kaluza is navigation. Things like tabbed analyses, the radial menu, and 'office-like' ribbon menus make performing simple tasks inside the software, well, simple. There are a few things I'm a bit unclear of at this point. Apparently batching stats and pdfs are possible in the software, but i haven't quite figured that out yet. Perhaps now that it is out of beta, and user manuals are being put together, I'll be able to play around with that more. However, it is definitely not as powerful as the batching options in FlowJo (their words, not mine). Also, being from Coulter, the nomenclature of certain things is definitely derived from the clinical world, which takes some getting use to. Panels, Protocols, Tests, are all used in a strict clinical sense, and don't always translate directly to the research world, so figuring out what is what took a bit of effort on my part. I have a feeling, the 1st adopters will probably be Coulter hardware users and disgruntled FlowJo users. The speed of analysis is a tempting feature for potential converts, but I may hold off until version 1.x or maybe 2.0 and see what happens. Until then, if you're just dying to try something new, let me know. I'll probably load the software on one of our analysis computers so people can play around a bit if they'd like. Or, if you'd like to get a demo copy of your own, visit http://www.coulterflow.com/bciflow/kaluza.php and click on demo request.

UPDATE: (From Ernie Anderson, BC) FYI, batching stat and PDF output is pretty easy. Just load the files you want to process and select them in the analysis list. (The usual Windows SHIFT-click, CTRL-click, CTRL-A shortcuts work here.) When you have more than one analysis selected, the display is replaced by several buttons:

"Export statistics from selected" will produce a single file containing the displayed statistics from each analysis. Make sure you're showing the stats you want to display before you export. If you're not picky it's easy to just select all the plots and turn on all the stats.

"Print all sheets from selected" and "Print report sheets from selected" do what they say. If you want PDF output, choose a PDF printer driver. There's one on the program CD if you don't have one already. (Sorry, you didn't get that with the beta.)

New "Online" Tools from Coulter... Almost Useful!

If you hop on over to the flow cytometry centric Beckman Coulter portal, aptly named coulterflow.com, you'll see a link to some online tools. Tip #1: These aren't actually online tools. They are downloadable, standalone programs which pull information from the coulter web site as needed. Tip #2: They're pretty crude at this point. I can definitely see the utility they may bring in the future, but as of today, they need a lot of work. So, what are they? One is a fluorescence spectrum viewer, called Spectrios (a little corny, but hey, they're rebranding themselves), and the other is an Experimental Design Tool called Experimental Design Tool, but I've renamed it to ExDT. Spectrios does pretty much what you'd expect it to, but again, it doesn't seem ready for prime-time yet. The good thing about Spectrios is that it's a standalone program, so you don't need an active internet connection to use it.

That being said, it's not nearly as good or as informative as the Invitrogen Spectra Viewer or the BD Spectrum Viewer (both online tools, though). Spectrios gives you the basics; Absorption curves, emission curves, laser lines, and filters. It does not calculate theoretical compensation nor allow you to add custom laser lines/filters. It is also lacking many fluorochromes from the list, but surely that will evolve over time. Spectrios and ExDT are so new and unpolished that they even forgot to take off the ***Concept - Beckman Coulter Confidential*** text in the header bar, and as is probably evident by now, these tools are still in Beta (v.0.9.0.7657).

Now here's where things get exciting. ExDT is a tool that will (once it's finished) greatly change the way you design your flow experiments. Built into the design tool are stock instrument configurations (laser lines, filters, # of detectors, etc...) and the entire Beckman Coulter catalog of Antibody/Fluor combinations. So, there are basically 4 steps to walk through (Screen shots below). Step 1 is a basic description of your experiment... Name, date, cell type, descriptions, etc. Step 2 allows you to select from the Beckman Coulter online catalog which markers you want to use. Now here's the fun part. As you go through and select which markers you want to use, the panel at the bottom shows you all the available Fluors for the markers, and tells you how many possible combinations can be put together to make your panel. Also, if you wanted to force one of your markers (say CD4) to definitely be PC7, for example, that will automatically narrow your possible combinations greatly. So, if you know that the CD4 PC7 is a really good antibody, and you definitely will use, just Pin the antibody to that fluor, and now you may go from 30 possible combinations to 6. Step 3 then, simply allows you to flip through the possible combination and choose your panel. As you flip through the combinations, you get a "Spectrios-like" window showing you the emission curves so you can determine which panel you like based on available filters, or reducing compensation. Does it rank the combinations? Not sure if it does now, but that'd be cool. Does it allow you to upload your own catalog of antibodies? Not sure, but can you say awesome! Does it allow you to input your own instrument configuration with laser lines and filters, and then pick antibody/fluor combos to maximize sensitivity and minimize overlap? No, but if it did, I may faint with excitement. Finally, Step 4 allow you to set up your run list; Single stain controls, FMO controls, Sample tubes, etc... Oh yeah, you can conveniently click the "Add to cart" button just in case you don't have those antibodies on hand and you need to purchase them from "you-know-who." I call that Marketing Genius! Once you've made your run list, you can print it out and send it to your local Flow Cytometry Guru to give his/her blessing, and away you go. This tool has so much potential, I really wish it was working to its fullest right now. I have all these ideas swirling around my head on how I could use this yesterday. I guess we just play the waiting game now.

Imaging Indifference

I can't really get excited about imaging data. I don't know why, they're always really pretty pictures, but today I think I figured it out. Here's how a talk including imaging data typically goes. Slide 1: here is cell line x stained with Nuclear Dye A, and here is the same cell line stained with Membrane Dye B. Slide 2: now let's look at those two images super-imposed. (Ooos and Ahhhs radiate from the crowd). What do you do with that? How does this translate into some sort of metric than can be quantified and then objectively compared to a different situation or cell line. Flow Cytometry, on the other hand, is very quantitative, easily comparable, and believe it or not, objective. I understand the value of microscopy images, but I always feel like it should be validated and quantitated using some analogous technology like Flow Cytometry. Sorry Imaging folks... You may be asking yourself, well, if you're indifferent to imaging, why are you getting an ImageStream? Aha! Herein lies the power of the ImageStream. Pictures for the picture people, and quantitative data for the quantitative people.

Pre-GLIIFCA Core Manager Meeting

Well, another successful Core Manager Meeting just wrapped up here in PIttsburgh, PA. Topics presented ranged from Core Facility management, and cost recovery, to training issues, SOPs, and writing effective SIGs. Many of the talks recapped events from the NCRR/NIH Workshop in mid July. The overall meeting was entitled "Efficiency and Quality in the Management of Core Facilities: Best Practices for Running Your Core as a Business." I can honestly say, this Core Managers' Workshop provided some of the most useful information I've seen in years. It was great to hear many other core facilities around the country (and world...since we did have one attendee from Denmark) run into the same issues we do at the University of Chicago. We got some great ideas that we'll reevaluate once we get back home and see if we can implement a few of those things

Sort Cancelation Policy

I started sending out emails stating the following:

Please be aware that the Flow Cytometry Facility will begin charging 1/2 of the originally reserved time for sorts that are canceled less than 24 hours in advance. Last minute cancelations result in lost revenue for the facility since it is highly unlikely someone will be able to occupy that freed time slot. We certainly understand when experiments go bad and sorting is not possible, and in those instances, we'd prefer that you call/email us to cancel rather than simply canceling your time online. Our current effort is aimed at penalizing those abusing the self-scheduling privileges. Please consider this as a friendly notification.

If you receive an email like this, it is because you canceled a sort reservation less than 24 hours in advance of the sort start time. Despite our efforts to get people to schedule sorts responsibly, on average, we're having 3 last minute cancellations per week. When these people cancel, they delete their time on the scheduling system anywhere between 30 minutes to 4 hours prior to the planned sort start time. What’s even more frustrating is that it seems to be the people who book large time slots (4+ hours) that tend to cancel last minute. So as much as 12-15 hours of prime sorting time can go unused per week. Of course, this is especially frustrating since there is a substantial wait time for sorting in the afternoon.

Now, as I say in the email, we certainly understand things go wrong, and will definitely give you the benefit of the doubt, but instead of just going to the online scheduler to cancel your sort, it is imperative to get a hold of someone in the flow lab and tell them personally that you will not use your sort time. Many times, we are spending 20 minutes setting up the instrument specifically for your sort...changing the tip, putting in the proper filters, aligning the appropriate lasers, etc... Also, we are sometimes aware of people who are 'standing by' just in case someone does cancel. If you can let us know immediately when your experiment goes awry so we can try to get someone scheduled for that time slot, we would greatly appreciate it.

How can you get a hold of us? Well, there are many ways, but here are the best ways.

• Call the lab at 2-9212
• send us an Instant Message at 'flowhelp'
• Send an email to ucflow'at'gmail'dot'com
• stop by the lab in Kovler 037

Keeping track of last minute cancellations is a real pain, and I hate doing it, so please try to schedule your sort time responsibly so I don't have to do this anymore. Remember, schedule what you need, use what you schedule. Thank you.

FlowJo Training

Thanks to all who attended the large and small-group FlowJo training sessions on September 16. It was very successful, and we do intend to repeat this at least yearly. Can't wait that long, you say? Well, FlowJo offers 'intro' type online training session on the 1st Friday of every month at 12PM C.S.T. All you need to do is at 11:55AM on the 1st friday of the month, click on the following link, sign-in, and you'll receive a toll-free number so you can listen-in as you watch the demonstration from your own computer. Here is the link: https://treestar.webex.com/treestar/j.php?ED=109322642&UID=1022933222

Additionally, the flow facility will train you on-demand 5-days a week, from 9AM - 5PM. You can schedule this by calling the facility at 2-9212, or sending an email ucflow@gmail.com. We're also throwing around the idea of having FlowJo open office hours once a week that you can bring down your data analysis questions, and get some personalized help. For example, we could set up a 2-hour time slot on Tuesday afternoons, and be available for any FlowJo questions you may have. If this sounds like a good idea, be sure to let us know.