Wednesday, July 15, 2020

Nanospray (&Microspray) sources on the cheap -- build your own!

I brazenly stole this photo off Twitter this morning from Ryan Julian (JLab @ University of California Riverside, lab website here).

Recognize part of it? Did you throw a couple out? Hold onto it! They came up with a brilliant way of utilizing it for good.

On that topic, every year or two I remember to redirect people to the amazing UWPR Resources for all things proteomics.

For years, UWPR has kept an awesome part list and directions for how to make your own nanoflow source for Thermo instruments -- both the traditional LTQ interface (that is also on the Q Exactives) and the newer NG source configurations.

I hate nanospray, but my plans for single cell for 2020 aren't going to go very far without it, but it doesn't take too much imagination to rig these up for Microflow/microspray, (that fine line before you get <1uL/min when everything gets much harder to plumb and do correctly).

You can run lower flow rates on an IonMax or HESI II source with a smaller diameter capillary (old post on that here) which is great for me when I use an S-Trap mini cause I'm starting at 100ug of protein or more.  If I'm down at the S-Trap might be time to start thinking about where I left the EasyNLC..... OR  I can work on assembling a MicroSpray source that doesn't shoot my peptides at such an extreme angle. That sounds like a lot more fun!

Tuesday, July 14, 2020

OCCPR Webinar Tomorrow (7/15/20) Alexey and MSFragger!

If you are going to catch one random proteomics webinar this week, may I suggest the OCCPR Webinar tomorrow?  You can sign up (free!) here (you do need to register and that takes a few minutes).

There are many mysteries here and I think this webinar will make an attempt at answering some of them.

On the MSFragger software:
How is it so crazy fast?
Seriously, what is it doing that other programs aren't doing?
What's next for it, and the Dark Proteome it was designed to explore?

There are other mysteries -- what about the man behind the team behind it? Alexey Nesvizhskii?
He's done a lot for proteomics. Is he a Saint?
His dabblings with diverse subjects such as LuciPHOr , odorous things, and the Abacus suggests he must be a Philosopher (new paper 7/15/2020!!)
I just know that when we really need this mysterious hero, we can put a symbol in the sky.
Finally, and perhaps, most importantly do you properly say his last name..... I've heard several valiant attempts.

These and other questions may (or may not) be addressed tomorrow. I'll see you there.

Monday, July 13, 2020

PEPPI-MS -- In gel separation of intact (and NATIVE??) proteins!

What if you could make top down (and NATIVE!) mass spectrometry a whole lot more accessible -- with gel based separation? Would that make you peppy?

The P stands for Passive and I'm not sure why, but if more than a few of us are ever (really) going to adopt top-down and native proteomics (I don't mean super pure antibodies, I mean the complex stuff) we need to simplify the front end and this looks like a great new tool for it!

You can check it out here! (Open access!)

Sunday, July 12, 2020

New and improved tools for the Omics Crew!

I'm a couple weeks behind...on...well....everything....but I really dig July's special edition cover -- and, in particular, this great summary piece!

Want a summary of some new -- or newly improved tools for your workflow? Here are 37 new ones organized out! 

If nothing else, bookmark it, and the next time someone comes by your office with something that you feel like you could probably pull off, but aren't sure how with your typical workflows -- maybe it's here!

The summary is called --

-- and if you see a technique on the list that you've never ever heard of -- it's probably in this special edition!

Someone remind me to read up on block's not in my folder, and I think I've retweeted it 4 times so I'll remember....

Saturday, July 11, 2020

ProVision -- ShinyApp for Quick Analysis of MaxQuant Output!

"....Google Images, making weird mass spectrometry blogs even weirder, since 2009.." 

I'm running a lot of MaxQuant right now because I'm running lots of really complex samples with short BoxCar runs and -- if you to quantify BoxCar data, MaxQuant has a little box where you put a checkmark and then it does magic. There are 2 other options for processing BoxCar data.

1) Gibberish quan

(Please correct me if something has changed recently, but as long as MaxQuant works, I'm honestly cool with it for these projects)

For MaxQuant output, you have loads of options.
1) An persistently hideous spreadsheet format that Excel has even more persistently refused to open 100% successsfully for nearly a decade.
2) The ultra powerful, can do absolutely everything if you're smart enough, will do nothing whatsoever if you are not quite smart enough, constantly improving for people smart enough to run it, constantly infuriating for people not smart enough to run it the end of the day....some of us aren't getting smarter....that train has sailed....

There are other options out there that try to bridge this gap -- and here is a new web-based one (which was what I meant to type about, forgot, and then the espresso finally got to my brain, 

You can read the paper (it's open access!) or you can just hop over and try out the program online here.

Now -- it is Shiny -- so, like all R/Shiny programs it is very particular about what your rows/columns are named. There is a function in the app to rename them, but I can't get it to work. I can only proceed if I go into my proteingroups.txt and rename them there first.

What do you get from it? Fast and pretty output. (Example plots from paper shown)

There is a funny number of warnings in the online version about imputing missing values, probably an appropriate number of them, which gives you a feel for the authors thoughts on the topic.

You can also run it locally through R in case some weird blogger is uploading tons of stuff to it and it's running really really slow today. You can get it here.

Friday, July 10, 2020

FeMS Transition Talks! Submissions open!

FeMS is at it again, coming up with innovative ways to get interesting science out that we might not otherwise end up seeing -- despite the challenges of 2020.

If you're near the end of a stage in your career -- or know someone who might be interested in showing some cool stuff to people at the ultracool Females in Mass Spectrometry organization, send them to this link! 

Thursday, July 9, 2020

Make sense of what all these new glycopeptide software packages do with this new review!

This brand new review at MCP is a tremendous resource. And someone is very good at making figures.... 

By separating out the different types of software into distinct categories it helps you to figure out what you actually need based on what you want to do with the data you've acquired.

I won't lie and say I had time to read this, but its hear so I'll remember to read it later.

Wednesday, July 8, 2020

MSFragger tears through TIMMYTOFF data at a process each 2 hour run!

MSFragger is FAST. Like really really fast. The original command line version could finish processing a Q Exactive file in about the amount of time it took me to push the enter button. As features have piled up on MSFragger to make it more useful, it has slowed down, but it is still a program that is darned fast for a desktop CPU program. says something....I'm not sure what....something, for sure....when a 2 hours TIMMYTOF run with PASEF takes 70 minutes, on average, to process with MSFragger and that is massively outstripping the competition.

Good news -- there is another tool to process PASEF-TOF data with!
Bad news -- you still have to be patient. This data is dense!

You can check all this out here.

Friday, July 3, 2020

What's an "instrument qualification" and should you think about getting one?

Whoa. This blog has suffered due to the fact that I'm just tired of writing. Or just tired. One of those. My last shift in clinical chemistry was over a dozen years ago and while the instruments have gotten much much much better, the amount of paperwork necessary to prove that your assays are valid when it really counts has....gotten better if you are thinking about thoroughness (the most important part) and...hmmm.... nah...not sure where I was going with that. I'm sleepy even though I'm staying within walking distance of my lab so I can be here ALL THE TIME! WOOOOOHOOOO!

What I thought would be interesting to ramble about over lunch is INSTRUMENT QUALIFICATION (sounds the oppositive of interesting, right? However --did you know that when you get a new instrument that you can purchase a whole ton of extra installation things to verify that your instrument does everything that you assume that it should?

It's called an "Instrument Qualification Package" and basically every vendor offers these. If you aren't in a clinic or some other regulated environment, your sales rep might not bring it up, but I think it is something everyone should at least consider with their new instrument purchase:

Why would you do this? A bunch of reasons
1) You get a bunch of stickers that are signed by a special Field Service Engineer who is approved to check all these aspects of your instrument
2) You get this special FSE around for days. After my installs I got 2-4 extra days of onsite time with an FSE who reeeeeeaaaaaally knows that instrument inside and out.
3) You get hundreds of pages of extra documentation on your instrument and its specifications:

This is just the extra documentation on the Q Exactive. What's in it? Tons of great information! 

What about a pressure test of your LCMS system and it's linearity? Do you get that during a regular install?

(Maybe they do it, but I've never gotten a hand signed report and comparison to factory data!)

Sensitivity metrics compared to factory specifications? Heck yes, you get that too!  My new QE has a really great S/N vs spec (due, primarily to low noise levels, which is likely temporary, but how cool is it to know that?)

Even cooler, maybe, is the fact that you get your software installed and pressure tested!

This might only be for targeted or for EFS/Clinical stuff, but I'm not sure. The FSEs brought their own data and processed it through the versions of software that they installed to verify 1) the software all works right and 2) it produces the data that it should AND 3) they left the data. You can quickly master the software by trying to replicate their results. How often have you wished you had a good file set for learning your new instrument software with? Turns out you can just buy that.

On top of all of this -- there are certain lab certification processes where all this information is required. Without it, you can't get contracts or jobs for some government agencies, etc.,

Worth noting, this is not free. For 3 instruments this was around $30,000 USD extra, something that I was initally annoyed about --- but I got close to 12 extra days to annoy some of the vendor's top engineers with days of uninformed questions about the inner workings of these boxes. And they installed all my software and gave me over 1,000 pages of paper, much of which they personally signed. In the end, I'll definitely do it again. Maybe soon, cause the new Exploris systems are surprisingly affordable (no joke, if you're thinking about any mass spec this year you should get some new Exploris quotes, you can score one [with a D20!] for less than most vendor triple quads.'re giving up some functionality as the number on the front of the box decreases....but I'll take a quad-Orbi that has recently been on actual fire (or possibly currently is on actual fire) over most unit resolution instruments, but if you've had the misfortune of being on this blog much, you already knew that)