Real time search allows for near perfect SPS selection!

I "play" a decent amount of basketball. It's "play" because...well...I go suuuuper early in the morning for pickup games because it's typically other old guys doing their best to avoid high school kids or people in their 20s who can still actually play. Sometimes -- maybe just because I need an ego check or something -- I hit a park or YMCA around Baltimore during daylight. That's pretty good for establishing my normal baseline level of humility for 6 months or so.

The way I feel leaving a Baltimore YMCA at noon? That's sometimes how I feel in this field. I really like our preprint where we quantified the SPS accuracy in some old datasets. It was a really fun study and after some advice from Phil Wilmarth I think it'll probably get published somewhere.

Okay -- so....what if instead of measuring that accuracy after your experiment -you ON THE FLY checked the SPS fragments to verify that they're from your peptide of interest -- and only used the ones that are right?!?

You'd have this preprint from the Gygi lab.

%SPSAccuracy near 100%....man....okay...so SPS MS3 is still slower than MS2 based labeled quan, but if you can make that level of improvement in the accuracy...?...

I hope this option rolls out for everyone out there with a Tribrid! If you're doing TMT you should 100% definitely check this paper out.

New MetaMorpheus Updates for version 301!

MetaMorpheus just got better again!  Man -- I love this autoupdater thing.

Go to launch my favorite search engine and --

"Hey weird blogger guy!  Would you like to massively upgrade this free software you use all the time? Here are 10 reasons for why you should just hit this button and watch another 2 minutes of Dunk League while I install cool free stuff for you."

"...let's get it poppin!..."

ASMS 2019 -- Maybe my final wrapup?

Maybe this will be my last post on the magical event that was ASMS Atlanta. To be honest, I came to ASMS to do a lot of boring things -- or at least not really proteomics things -- like small molecule instrument shopping (ummm... if you haven't seen it...that thing Conor is holding in the picture above is a fricking ion trap. A surprisingly high resolution ion trap that we demo'ed at a secret location (which made it even cooler! I was frantically texting my team that we all had to meet at these elevators before they closed and we all made it and were rewarded a fully functional with MX908 we could hold and test things on!) We had some problems during the demo because we kept underestimating how sensitive it was and the little thing would initiate a 2 minute self-cleaning procedure -- cause that's a thing that an ion trap can do?!?!?) See -- boring stuff -- but, -omics mass spec isn't the only thing that has been advancing like crazy.

One really big thing for us for ASMS was -- okay -- for real -- how do we take this RNASeq and PacBio data and integrate it into proteomics? While running to a meeting we scored with Dr. Jack Henion -- which I was NOT going to be late for (he's awesome btw, and ---

-- can confirm...) And...if you haven't seen his awesome little mass spec you can juuuuust fit into the hatchback of a 2011 Chevy Volt and can power for a full day off of it's internal battery (I have done all this math myself, which means it's probably wrong, you should check out this dramatization)

-- Maybe found my favorite thing of the whole conference. David Tabb with a microphone and a projector right in the middle of the session hall enthusiastically explaining how to integrate "next-gen" sequencing data with proteomics -- 

-- it seemed like the most random thing ever -- like David heard everyone in the US was struggling with this, so he flew over with his own projector and a megaphone just to help us (it was slightly less cool than this -- it was a new-to-me session format of informatics talks during the poster sessions and I hope it never goes away. There was a great talk from Megan Burke on NIST Hybrid Search the final day I only got to catch for a minute. ProSIT+Hybrid search time?)

OH. THIS. Should be mentioned.   MSQC.LIVE  It's still in beta, but Ben Neely gave a great talk about what it will be able to do in order to make proteomics QC super easy to interrogate. It's worth mentioning that this is an unsupported project. The people building this are just QC nerds doing it in their spare time. 

I think I saw 3 applications of "chromatogram libraries" -- and now I think I get the concept. It's been on the blog before, but I think I mischaracterized it. You should ignore anything I've written about it and just look at this cool paper. 

Yeah...I should probably wrap this up. It's gotten too rambly and chaotic. I was explaining to a collaborator at the NIH who was attending his first ASMS one of the big challenges here that I don't face at other meetings I go to. At ASMS people show off new stuff. Hot-off-the-press stuff. Poster abstracts they submitted in January and the experiment didn't actually work until the day they were running out the door to get on the plane stuff. That seems normal if ASMS is your conference you attend all the time. This isn't normal in general. People don't go to AACR, for example, and show data they haven't published yet. They talk about the stuff they've already got securely in press. They keep their data close to their chests and in locked freezers back home and they keep their lips sealed so they don't get scooped. I've got loads I can't wait to revisit later.

That's it. Done poorly writing about ASMS. I'm already looking forward to Houston. Though... I am trying to figure out how to justify going to Adelaide first....I've almost got the airline miles to pull it off...

Oh...and I also got my Ocean Protein Portal sticker back (had to return my old work laptop and couldn't get the sticker off in one piece-- and I got a bonus sticker that gets me odd looks when working in airports....)

RforMassSpectrometry!!

Okay -- R can be a little daunting because of just how many packages and resources are out there -- but what if someone really went above and beyond and organized all the things that I actually want to see and use?

Then you'd have this stunning website!  www.RforMassSpectrometry.org

ASMS Recap Part 3 -- Sunday and SKYLINE USER MEETING!

Correction: Sunday was a big day. I've already mentioned equipment stuff, including the FlexTimsTOF,  the Fusion 3 system and the Exploris 480. I've had trouble remembering the name of the last one and heard other people have as well, so I came up with this image to help me. 

(The 480 is cause that's the resolution the little thing can do!) 

There was new hardware everywhere and awesome vendor events on Saturday, but Skyline was what I was looking forward to. 

The Skyline venue was great (Georgia Aquarium) And packed! There were people standing/ sitting. And it had a really cool place for coffee  breaks.



It was the 10th Anniversary of Skyline and we heard from many of the developers and the first users of the software that helped shape it toward being something that is activated thousands of times each day around the world.

Again this year, the organizers tried to pick talks that really demonstrate the flexibility of Skyline to adapt to new applications.

This was the schedule (you can click to expand it)

New applications I really took note of?

Skyline for nucleotide quantification studies!  This is sooo cool because along with all these hardware advances that have helped us on the protein side, the hardware is getting pretty good at the nucleic acid stuff as well -- and here is your data analysis package. You already know how to use it!

Using Skyline for GC-MS data?  This is reaaaally smart. GC-MS ionization is mostly rough ionization, so you have loads of fragment ions and often very little of your parent ion around. Sounds a little bit like DIA, right? That's how Pawel Sadowski's lab is doing GC-MS based metabolomics! By telling Skyline it's metabolomics DIA data. He also showed how you can deconvolute overlapping peaks (in GC-MS this is a very different concept than in intact proteomics) within Skyline itself.

...and...then...this....happened...you had to be there.

Then we got to see how PROSIT can generate spectra for Skyline and how well it performs compared to previous in silico fragment generation data. Super slick.

We're eventually going to do global untargeted studies on everything on this planet. When it comes to more focused studies guided by all that stuff, it's great to see that 10 years of development has provided us a free tool that can take us to the next step in targeted validation of both new and historic data.

JASMS ACS Publishing Partnership

JASMS will be published by ACS going forward. Credit goes to MassSpecPro, whoever he is, for finding this.

ASMS 2019 Recap Part 2 -- Saturday's Power user meeting!

Exactly one week late! Time to recap this awesome meeting. If they do this next year, I highly recommend you check it out. Shannon assembled an amazing workshop. If you've been doing this for years and think you know it all -- this is where you learn something new. It was packed. They definitely have to cut off attendance -- the best part might be how engaged a smaller group could be with questions/answers. There were some smart mofos in the room.

Agenda (mostly accurate, small change --it's on Planet Orbitrap here or you can click to expand):

My highlights:

Calibration and Preventative Maintenance -- 2 real life service engineers, Faith Robison and Kathy walked through user maintenance stuff in detail. Boring? Yes. Okay -- but did you know you should ballast the Sogevac roughing pumps? The Edwards? Sure, everybody knows that. But I didn't know the Sogevacs need ballasted! WHAT? Ummm...I will probably just pretend that I actually have known that and should delete a few sentences here for purposes of my own credibility.  I'll try to remember to link those slides when they're up on Planet Orbitrap. We also learned how to clean the Exactive family side filters and why you should do it. Important stuff in there, for real!

FLEXMIX!  If you're using your instrument for multiple things, chances are you waste a lot of time changing from positive calibration solution to negative, to something to calibrate your extremely low and extremely high mass ranges. Swap syringes, wait forever for bubbles to come out, start too early, spray still isn't stable. Wait some more. Repeat.

It's live for sale here.

It's ONE SOLUTION. No switching. It's got all sorts of stuff in it. It may only be natively supported at this point by Tribrids, but I plan to manually punch in all my calibration points and just use this going forward.

Maybe I'll let you know if that ends up being a really bad idea -- but it seems smart to me.

The awful terrible catastrophe of an operating system that is Windows 10 -- NO I DO NOT WANT TO USE BING -- IS IT USING BING TO SEARCH MY FILE FOLDERS? I COPY/PASTED THE NAME OF THIS DOCUMENT IT CAN'T FIND IT, WHY IS IT GOING TO THE INTERNET TO LOOK FOR MY WORD DOCUMENT?

Windows 10 Compatibility. There is one official version to use with the newest Foundation/Tune packages -- Windows 10 Enterprise LTSB 2016 -- and it sounds like Exactive 2.11 and Fusion 3.1(?) forward are approved (always check your software release notes before installing. I'm often wrong). My advise, ignore Win10 as long as you possibly can, but if you have to upgrade the vendor is thinking ahead.

Dr. Shannon Eliuk and new super powers for Fusion I and Fusion II Lumos -- I can't wait to go through these slides. They have, yet again, rolled out upgrades to the tribrids already out there. I can't even go into 1/10 of the cool stuff. I spent a lot of time slack jawed. My favorites (and I can't remember if each one is 3.1 or the upcoming(?) 3.3(?):
Multiplex with variable window control. BOOM.
Assisted HCD optimization (I think this was IDX/Lumos only, notes don't say)
Normalized AGC target support (for people who have both Exactive and Tribrids in their labs)
Precursor Fit filters -- You can preferentially select ions that match your theoretical isotope model over ones that don't
Intensity thresholds in exclusion/inclusion lists
Quality filters -- only do MS3 if there is plenty of signal for the ion you want to do it on. You don't waste all that time getting to MS5 fragmentation -- only go as far as it makes sense -- on the fly!
Advanced targeted inclusion -- high priority vs low priority list -- I've already designed an experiment around this and can't wait to see if it works!

Dr. Tara Schroeder and ALMANAC -- Thermo is really focusing on lab efficiency. Remote monitoring and diagnostic info acquisition support. If you can get all this rigged up (use the Thermo Cloud Connection thing!) your local FSE can arrive with a set of diagnostic information that technical support has already reviewed so you get up and going faster.
Ooooh...And there is a LogBook functionality....can multiple people walk up to your instruments? Want to be able to go back and see who f-ed it up? There are logs! Finally make the powerpoint you've always needed so you can get rid of Steve! (Name chosen at random.)

Dr. Aaron Robottails (pronounced RobItI, I think) and pushing the tribrid to 8k human protein IDs in one shot with FAIMS. I'd love to get that method, I don't have anything for FAIMS. The highlights were narrow isolation for precursor (0.7Da) and using the ion trap in turbo scan. I'm a little iffy on this last one. My turboscans don't look pretty, but maybe that's because I'm not isolating narrow enough (and, without FAIMS) maybe with 2 ways to eliminate isolation interference turboscan looks great?

INSTRUMENT CONTROL -- What many of us came for!  I'm embarrassed that I was so excited I didn't catch the two great R&D guy's names. Wait. Found a note. Jesse was one speaker.

To keep in mind. There is an XML method editor (this is for taking your Fusion method and either automating it when it's super complex, or building something new inside of it.)

This is different from the iAPI, which is actually performing external thingies like taking data out and putting data in. I'm no programmer -- I brought an awesome with me -- but they've done a great job of making this stuff seem approachable. They have a GitHub here with great details on how you get this stuff and how you get permission to use it and everything.

 I'm going to cut this short. My grass is like 1m tall.
There were more great talks and mostly customers.
Dr. Florian Meier showed MaxQuant.Live (the 2019 version has been downloaded 500 times!)
Dr. Chris Rose (now at Genentech?) demonstrated that all of us normal people can now use TomaHaQ thanks to the power of the TomaHaQ Companion (thanks for the shoutout, Chris!)( also -- Simion Kreimer just provided me with some pre-made methods to make publicly available -- I'll try to get those up soon)

And I missed the smart TMT and small molecule talks for another meeting. I can't wait to see the slides!

Seriously, thanks to the vendor (and especially Shannon) for putting together such a great day. The most sincere compliment I can pay to anything is to be somewhere for that anything at 8AM. And I was there chugging coffee and pretending to be a sentient organism. At least 2 people can verify that.

ASMS 2019 -- My first pass super sleepy summary.

 I obsessively take notes. And then leave those notes in stupid places. Now, thanks to the power of the Cloud things my notes are on electronic devices and they're all linked all over the place, I lose them less rarely. My notes for ASMS are nuts. Like 50+ pages of things that look like that craziness to the right. I might not get through all of them in a year (it's happened before...) but here are my big takeaways from this year's mass spectrometry get together.

1) On the hardware side we might have finally gotten past the exponential growth phase in instrument performance. Yes -- for sure -- instruments are better this year. The Eclipse definitely deserves to be the Fusion 3. It's a step up. The 480 is the coolest benchtop Orbitrap we've ever seen. But -- the big advances aren't the introduction of the reflectron or cutting the diameter of the Orbitrap in half, they're primarily taking these pieces we have and putting them together more efficiently -- and -- this is big these days -- in making much much smarter instrument software! 

I don't mean to insult any vendors by this. The MALDI on the TIMSTOF? That's sick. Imaging mass spectrometry got a big level up.

People seem excited about the cyclic ion mobility on the Waters instruments. I still don't understand the ramifications of this, but maybe if non-cyclic (?) ion mobility wasn't doing whatever it is that you use ion mobility for and there is some reason to believe this is the right technology to do it -- maybe you need cyclic. 

But -- by far the biggest things I saw and heard were -- intelligent acquisition strategies and on the fly decision making and what really looks like the vendors taking feedback from the MS community about what craziness they need and helping them enable it. 

2) TOP DOWN PROTEOMICS might really be a thing soon. There were posters everywhere. There is awesome new free user friendly software (MASH Explorer -- coming soon!?! -- haha! I'm a beta tester so I can use it.)

FLASHDeconv is available for download -- yo -- ever thought that Xtract was pretty good at deconvoluting things, but wished it was 1,000 times faster?  Respect is pretty fast -- but it's got that weird duplication thing (it sometimes can't tell if your protein is 22kDa or 44kDa or 66kDa, cause, in the end the algebra cancels out the same in isotopically unresolved high charge spectra). FLASH is faster than Respect without the duplication stuff --  it operates in OpenMS -- so if you're already a user you've got the interface down! 

3) The second party companies for hardware and software are getting increasingly sophisticated. There were multiple booths with ways to upgrade your existing hardware boxes. OmniTrap was there as well as the FTMSBooster and the crazy thing where the microscope time of flight and Q Exactive are connected and function in unison.  Cool new sources and separations and 200cm nanoLC chips with low back pressure (with a new US side distribution partner)

4) You don't seem like a weirdo anymore if you talk about proteomics quality control! Everyone is doing quality control! Yes. QC is boring. QA is boring, but this intimidatingly bright next generation of mass spectrometrists I met everywhere are going to exceed what we've done here in every conceivable way and I'd almost think the majority have QC as a mandatory step in every experiment. I'm pumped. 

5) 

Okay. I don't know if that worked. It's funnier to me if it didn't...."say hello...to the robots..."

Who isn't pumped to get automated?!?  Exploited people Postdocs and grad students are awesome and everything -- but they're way more awesome when they can use their brains with all that enviable neuroplasticity to come up with smart new ideas on how to do things and aren't doing dumb stuff that a robot can do better, like pipetting or tip based enrichment. There must have been 15 different robotics companies there looking to make mass spectrometry better! It's awesome that so much stuff exists now that is specifically designed for our applications.

6) PROSIT. PROSIT. PROSIT.  I ran into a member of this team and he misguessed how many talks his team had. They had a ton. Because there is so much you can do with it. 

7) There was another one I thought was critical, but I'm also very very tired for some reason.... 

More later, maybe, when I actually read some of these stupid notes!! 

I'm already looking forward to the next ASMS (Houston?) but I'm pretty glad I get a year to recover. 

A brand new Orbitrap has appeared! The EXPLORIS 480!

Okay -- so -- no one can possibly be mad at me this year at ASMS!  Details have been on Google images for 2 days and it was standing room only this morning for 8am reveals of the brand new Orbitrap system!  Hundreds of people know about this.

If you want details, there is a full spec sheet on the manufacturer website here.

Highlights?

Its an Orbitrap system that is about the same size as some of my HPLC stacks!
480,000 resolution!
40 Hz
Ion funneled
Mass range extended both lower (40 m/z!) and higher (up to 8,000 m/z with BioPharma option)
FAIMS compatible!
Smart logic stuff (I'll get more details, but it looks like it runs through the Fusion type software so you can build ultrasmart experiments)
The polarity switching seems faster
Multiplex up to 20 ions!
Ion Routing Multipole (like the Fusions!)

Oh. And...well....there also appears to be a new FUSION SYSTEM!!  And the number of features, including a new gas phase ion-ion reaction source, that it is (or can be) equipped with are just crazy.

Unrelated...

Finally! More instrument methods for www.LCMSMethods.org!!

HAPPY ASMS EVERYONE!!! If you aren't here, don't worry, I am. And I plan for zero sleep and 400% productivity. As proof -- I gave myself a deadline. Today. For new ready-to-go instrument methods.

You can get them at www.LCMSmethods.org or you can get them directly from the GitHub account here. (Click on the ASMS 2019 Update Zip file and it will take you to a page with a download button.)

What's in there?  How 'bout 107 new methods? That's a lie actually, many methods were just opened and checked for compatibility in Exactive Tune 2.11 to verify they worked properly. But there is a lot of stuff....like....

MaxQuant.Live is awesome, and once you use it a couple of times to get the hang of it you'll find it really easy to give your HF or HF-X some super powers. There is a folder containing both the Xcalibur and MaxQuant.Live methods properly linked. You'll just need to add your chromatography conditions and go.  I threw in the super speed method for the HF (run it at HF-X type speed -- but, be warned, due to the higher overhead, it doesn't perform as well as the HF-X). There are also some premade BoxCar methods.

Speaking of BoxCar stuff, I get loads of questions about running BoxCar like methods on the Fusion and Lumos. The Tribrid methods that come premade are AMAZING, so I haven't felt it necessary to release stuff for these systems, but I've thrown in my BoxCar iterations (built in Lumos with tune 3.1). I've also included some notes on the methods and things to help make sense of it.  Here are my definitions, though

BoxFahrt -- kinda like BoxCar, but not quite as powerful as MaxQuant.Live BoxCar

BoxCar Assisted MS/MS Fragmentation (BAMF) -- like BoxCar, but the ions selected for MS/MS are selected FROM the BoxCar scans rather than the MS1.

BoxCarDIA -- Utilizing the ion trap to do small DIA windows while the slow BoxCar scans are happening. I have files, but I haven't found any way to process the data. Want to give it a shot? Shoot me an email. I'll get you files from this Lumos I get to play with! My goal is to use the low res DIA for confirmation only (similar to WiSIMDIA)

Methods for the Exactives without Quads -- (Exactive Plus and Plus EMR) -- including multi-"phase" methods I've been developing on mine in Columbia. Proteins can be dumb about ionizing. One antibody might need 50 SID to ionize and the next 80 EV. By utilizing the "advantage"(?) of wide peaks for intact proteins, you can have multiple methods with different methods in a single peak and more rapidly figure out the ideal settings for that protein.

I'm doing a lot of work with pesticides these days -- my daily method for rapid pesticide screening with confirmation MS/MS for QE Classic.

What didn't make the cut?   Triple quad methods (I've got a TON of good ones, particularly for metabolites and pesticides!), Peptidomics. Lipids. Crosslinking on the Q Exactives, but it's some progress, at least!