Sunday, December 30, 2018

ABRF San Antonio -- Making a Proteomics Core Nimble & Efficient!

Alrighty Buckaroos. Time to throw another shrimp on the barbie(?) and mosey on down to San Antonio! for --

The "at a glance" is now up and it looks like another ridiculously awesome and action packed meeting for us core lab folk.

What I want to direct you to if you're making plans to attend is one of Sunday's workshops:

Making a Proteomics Core Lab Nimble & Efficient!! 

Let's all sit down and try and tackle the biggest challenges that face getting the turn-around and sample quality up!

Speakers include:

Brett Phinney ....who, while new to this core lab thing...wait...

...nevermind ---who will be talking about how to handle diverse (weird) complex samples within a core environment. (I copied that Tweet months ago and figured I'd use it sometime!)

Birgit Schilling  -- who will focus on implementing that DIA stuff in a core environment. (Oh -- if you haven't seen this new paper yet, I highly recommend it!)

and this weird blogger (Me!) -- who will try to say something coherent about addressing the key bottlenecks in a core environment! How do we wrangle up the biggest do whatever it is you do after you've wrangled something up....(giddyap? is that a thing?)

Giddyap, it is! Get that thing that arrived proteomerized or metabolomerized and get that data out the door!

Edit...misspelled "proteomics" in post title...

Saturday, December 29, 2018

HLA Splicing, Yes/No/How much? -- Part 17!


Are you working on endogenous peptides?  Are the awful fragmentation patterns of peptides lacking basic residues things you see for just a second when you first close your eyes? Would you LOVE to see another entry into the "are these things spliced or do they ionize and fragment so poorly we have no idea what we're looking at" arena?

You're in luck!

HLA peptides are BIG right now (for great reasons). And -- honestly -- what we know about these endogenous peptides seems to border right on nothing -- despite the fact Don Hunt's lab showed us we can study these things by mass spectrometry -- 25? almost 30? Years ago?  I swear the first paper is 1989, but Google Scholar thinks I'm making that up.

Friday, December 28, 2018

Biostars...always helpful...

Top comment on a question I was going to ask Biostars, but someone else already did....

...for some reason I had this weird idea that this was an okay place to ask questions...maybe cause this is where Bioinformatics is Explained...

I'm only miffed, because that was the only comment that wasn't an insult....

Washington Post Editorial if you haven't seen it already.....

Washington Post direct article is available here.

I'm sure they weren't trying to imply anything with the image. It was probably the only science picture they could find.....

Thursday, December 27, 2018

Hey Galaxy People! Want easy proteo-transcriptomics!??!

Edit: Forgot paper link. Here!

Galaxy is one of the biggest players in the genomics/transcriptomics playground. As I understand it (from listening to smart friends in these fields):

1) Loads and loads of tools
2) You don't have to be a super coding expert
3) You don't get access to super cool new code until someone integrates it, but the community is big and active enough that the only people that get access to new tools faster -- are the super coding experts

Galaxy for proteomics (often referred to as GalaxyP) hasn't been as ubiquitous in our realm -- we've got our camps like TPP, MaxQuant, PD, CompOmics, OpenMS, etc.,.

GalaxyP shouldn't be ignored, though, it's an amazing tool for the (seemingly much more common) bioinformaticians coming from the genomics realm to jump right into ours.

QuanTP is a great new gap bridging tool! You can statistically correlate the transcript IDs with the proteomics data within this! What else can do this? Perseus -- I think, but you'd have to work all the data up into the right formats which would require loads of preprocessing from both sides. (I could be wrong here, of course.)  In Galaxy you'd already have all the transcript processing tools available and the proteomics tools.


....Yes. I felt very stupid while making this. And while posting it. But it's done now!

Wednesday, December 26, 2018

Q4SRM -- Fast check of all heavy labeled standards in SRM experiments!

To avoid my post Xmas cheer rambling -- you should just check out this cool new QC software for heavy labeled standards in SRM experiments.

You can get the Q4SRM software here!

It has a cool logo!

Rambling you should skip:

I have a tendency to give triple quadrupole (QQQ) instruments a hard time. It's tough to go back when you've seen the power of 3 decimal places at MS1 and MS/MS. SPECIFICITY!!

I've done a couple of head-to-heads and I'm 100% convinced that even a Q Exactive Classic can still achieve higher sensitivity than any QQQ on the planet (note: in complex matrices! if you're diluting resperine in water, the QQQ wins every time. If you're studying pure reserpine go with the QQQ!!! 😇), but you need at least 100 ms of fill time to beat something like an Altis or a 6500+. That's -- at best -- 10 scans/second?.... our Altis can do over 600 scans/second... so....there are some definite advantages.

However -- SRM specificity still sucks. Wait! I read this review to help me sleep on a plane, but it ended up being awesome. I think it was this one (sorry -- its ElfSeverer.....) and the author brought up this great point -- the increase in sensitivity of QQQ devices has actually made the specificity problem worse. That when you can see lower and lower levels the chances that you will find something with an MS1 +/- 0.7 Da that also has an ion with an MS/MS fragment (or coeluting compound has an MS/MS fragment) within +/-0.7 Da gets increasingly higher! Interesting, right? Don't surplus that Quantum or 4500 just yet! 

Okay -- what was I rambling about (sorry...I should put a warning at the top...).

Q4SRM!  Okay -- the smart people at PNNL do a lot of specific targeted assays.

They do it this way:

1) Relying heavily on chromatography (what?? Since when does PNNL have an emphasis on chromatography being a critical component of proteomics? Where were they when we were all filling up TRANCHE with the worst chromatography of the last half century? Wait....oh...right...)
2) Putting in stable heavy labels for every peptide they quantify
3) Developing awesome software that can rapidly run QC on their stable heavy peptides as soon as the instrument has finished acquiring the spectra!

If you're going to do SRMs and you do those 3 things, I'm going to be super impressed. You've got an assay that is awesome!!

Get an SOP and get that thing out there helping people!

Is there other software you can make do what Q4SRM does? Probably! But we're all using different software packages and this is the first GUI I've ever downloaded that is specifically designed with this in mind. As light and fast as this is, you could stick to your normal data processing pipeline for your targeted quan and just keep Q4SRM on the acquisition PC to verify that everything is going according to plan.

BOOM. Massive point of data acquisition Quality Control upgrade!!  And that is NEVER a bad thing.

Tuesday, December 25, 2018

Happy Holidays!!

Happy holidays y’all!

This is the end of a scarf my badass sister knitted for me around her easy schedule of medical school at some famous university in NYC....wait...she also made this?!?!

Monday, December 24, 2018

DIWA -- Dual Isolation Width Acquisition for Combating Ratio Suppression!

We're not done innovating, people! Smarter instrument methods keep coming in all the time, and DIWA is an awesome and elegant new example that you can read about here!

Also, an earlier pre-print version is available at biorXiV here. I'm not sure how I missed it. It would have been really helpful...stupid job getting in the way of me reading every paper that comes out, I guess....

Here is the idea -- if you are doing MS2 based isobaric labeling quan (TMT/iTRAQ) you have 2 choices:

1) A really narrow isolation window that will help you get less coeluting ions. (Obviously of varying efficacy depending on your complexity, actual m/z, etc.,)

2) A wider window that will allow you to catch your more of your peptide isotopes.

-- here is a cartoon I have for something we're working on --

If you use a 0.4Da isolation width on any isotope of these two coeluting peptide species, you're going to miss a TON of signal. Maybe the majority of it! Are you going to get enough signal to get an MS/MS that you can match a peptide out of? Maybe....?

Dual Isolation Width Acquisition (DIWA) says --

(In case you're wondering why you remember this -- we did the research -- this was a commercial for a grocery kit that contained both hard and soft shell tacos. --@ProteomicsNews -- not afraid to ask the hard questions)

Two MS2 scans were acquired for each ion selected for fragmentation on an Orbitrap Velos system. In order to sort the scans out, they had slightly different HCD energies. A wider isolation was used to get the best signal for identification and super narrow isolation was used for quan.

How'd it do? Impressively well.

You don't have to believe the authors.  You can get the files at PRIDE here when the full print of the paper goes live

Things to note -- this was TMT 6 plex. I'll always hold the great Orbi Velos I once had in high regard, but it's slooow for TMT 10/11-plex. 6-plex? No problem!

This can totally be applied to the other instruments, like the 2x faster Elite or Q Exactives! I'm not sure how to write it for the QEs, though.... I'll think on it and if I figure it out, I'll add it. (Definitely ping me if you have the method!)

There are other important points here such as the ability the narrow isolation spectra give you to do regression analysis to assess your isolation interference. And -- this sounds totally smart -- the authors mention the potential application of ion mobility as something they might be exploring?!?!

Sunday, December 23, 2018

Prep for metabolomics AND proteomics out of the same sample!!

YES. THIS. This is exactly what I've been looking for.

Check out this ASMS poster from the great people at Cold Spring Harbor Labs! 

Sometimes proteomics will solve that tough biological question that smart biologist just brought you and trusted you to solve.

As metabolomics tools continue to mature, I'm finding just as many problems that can be solved with (much easier to set up from the instrument side!) getting a view of the global metabolome.

The problem has been that the classic sample prep methods force me to choose. Do I do proteomics? Or do I gear up for metabolomics? What if  I didn't have to choose at all and could get both samples SIMULTANEOUSLY?!?!?

Honestly, metabolomics run in triplicate sets me back about 45 minutes. It almost makes sense to do the metabolomics -- and if that doesn't seem to unlock the key to the puzzle then I've probably got digested peptides dried and ready to load. (This probably isn't the way I would do it since I don't have a dedicated metabolomics unit -- I steal borrow one that is meant for small molecule structural stuff when I need metabolics or top-down capabilities)

Either way, though -- the possibility that I could explore both avenues with one single sample?!?? That's a big deal for me and my collaborators.

Saturday, December 22, 2018

psims - A DECLARATIVE writer for mzML and mzIdentML!

As proteomics moves closer to the ideals of unified data formats (in both processed and unprocessed forms) we need to make sure every pipeline can support these formats.

psims is a great new resource for the weirdos who are using R and Python (!! joking!! just jealous of your neuroplasticity!!) to get into the uniform format bandwagon.

You can read about it (early/open) at MCP here!

If you don't want to read and just want the code, you can get it at GitHub here.

What makes it "declarative"? I don't know, but if the title didn't make you think of this, it will after you watch it.

Friday, December 21, 2018

A nucleic acid MS/MS search engine!!!!

Do you guys know how most labs check the accuracy of their PCR products?

By running gels....

It looks super precise -- until you realize that 100 base pairs is like 40kDa. You can get higher resolution (longer) gels, of course -- and there are other capillary based technologies -- but these things just generate an average mass. They don't tell you if something is wrong.

If only we had a NUCLEIC ACID SEARCH ENGINE!  We could use mass spectrometry to get a precise mass -- and sequence things! But that's more futurist mumbling, right?

Starting with BioPharma Finder 2.0 -- nucleotide deconvolution is a feature -- I was thrilled to find it and to see support in some other software as well. That's a good sign -- but a Nucleic Acid Search Engine is a critical next step.

If you'd asked me yesterday -- I'd guess 3-5 years and we'd see one...fortunately, I'm wrong a lot!

Consider how old-fashioned mass measurements are of amplified and often purified nucleotide sequences.

This group jumps a light year ahead and looks at how REAL RNA SEQUENCES are not only aligned -- but modified in an epigenetic sense.

Using HCD OT fragmentation of transfer RNA from people, they find they can monitor at least 20 different modifications -- effectively blowing open the doors on an entirely new way of studying epigenetics -- and maybe opening a completely new field for mass spectrometry!?!?!

I'm not sure the importance of this preprint can be overstated. This could revolutionize molecular biology and our understandings of the control between transcription and translation. And the method is straight-forward -- and the software is available for download here. Right now.

Thursday, December 20, 2018

Glyconnect -- the next big step in the maturation of glycoproteomics!

Okay -- this is a big step in the right direction. We're getting much much better at glycoproteomics as a field. I have a figure somewhere that shows how many studies y'all are knocking out year to year... seomewhere... and this is the fastest growing area. What we need next? This.

This could be the UniProt for glycoproteomics. The site is obviously well-designed and ready for incorporating tons of new data.

What it needs -- is tons of new data. Get on it, everyone!! 

Wednesday, December 19, 2018

3D Print custom labware!

I scored my niece a 3D printer for Xmas last year. Black Friday thing and a lower price than I'd ever seen one for. Unfortunately -- a little too hard for her, especially after the company went out of business and it had to be operated with a series of open source software in order to work.

After replacing it with a more kid-friendly App driven alternative, I end up with my own 3D printer. You can only print so many can print a lot of Daleks....but these things can actually be useful!

Did you know this existed?!?!  I didn't until just last week!
There is LOADs of custom labwear that you can easily download and print with glow-in-the-dark filament if you want!

A site called the "Thingiverse" is also loaded with great ideas like these.

I love all 3 of these. They're things designed to help those of us with freakishly large hands work with tiny PCR tubes.

There are some remarkable similarities between using the 3D printers and mass spectrometry. Resolution is a big deal. Expensive printers can achieve higher resolution in shorter time. Little discontinued $150 printers like mine can produce high resolution toys and tools, but it might take all night to do it.

If you've actually read this far and have access to one of these things -- this is the stuff  I've found the most user friendly:

Ultimaker Cura -- free software for taking premade objects and scaling them and adjusting them to print on anything

TinkerCad -- Create your own stuff for printing (which I then send to Cura and then convert to .gcode for printing)

Okay -- and just a reminder that THIS exists....

Monday, December 17, 2018

The flippening?

This has been coming for quite a while, right, but it still seems crazy. The Q Exactive hasn't been out all that long in comparison to our field -- but here we are. At some point since I last looked (can't find the post....but it didn't seem that long ago....) the Q Exactives overtook the great Orbi Velos for the most publicly available proteomics data sets.

And, at the rate that things are being published and dropped -- it won't be long before the QEs surprass all LTQ based Orbitrap systems.  The Elite is listed separate and is around 500 -- roughly where the Fusion systems are currently.

Sunday, December 16, 2018


Okay -- so TMT is great if you're comparing 11 things, right? But -- WOW -- does it ever get less fun to use when you've got 12 samples. Or 34.

There are many ways to tackle this, and these different ways have varying degrees of success. When in a pinch, I go to my dumb method, but it's VERY clear to me that this is crude and I'd love to have a better way.


24 (TWENTY-FOUR?!?!?) TMT 10-plex batches.

Coisolation explored.
Label isotope contamination explored.
Did y'all know that there is an entire chromosome(!??!) that roughly 50% of the human population has that the other half does not?!??
They use this evidence to determine the biological FDR. Cause...cells from humans that lack that chromosome don't have genes to make those proteins. If they show up in cells that don't have the genes, your FDR is probably askew....

The data is processed in MaxQuant. The post MaxQuant analysis was done with a bunch of R things, but the basic workflow is explained very well in nice figures and I feel like I can follow this logic to improve the next multi-batch set I try to combine.

Saturday, December 15, 2018

EuPA YPIC problem solved and published!

This challenge was such an awesome idea! And I'm thrilled to see that it was tackled and solved.

You can read about it here. 

Friday, December 14, 2018

TOMAHAQ Companion! Targeted mutliplexed proteomics for everyone!

Hey! Remember TOMAHAQ?!? It's the technique that Gygi lab developed for using internally triggered targeting for quan, kinda like IS-PRM, but then the thing you target is actually TMT labeled?!?

You can download the original TOMAHAQ paper from Molecular Cell here.

Okay -- so TOMAHAQ is suuuper smart. Unfortunately -- outside of Gygi lab, Rosa Viner, and Simion Kreimer -- I'm not sure anyone else ever used it. Why?

1) The experiment was nearly impossible to set up.
2) The data was nearly impossible to process.

What's on my desktop? Right now? 400 icons and a picture of Isaiah TomCat sitting on the head of a pug...AND...


Multiplexed internal standard triggered multiplexed targeted proteomics. For everyone. You can get the TOMAHAQ Companion software here. If you're paywalled for the paper, there is enough information within the GitHub download to set everything up.

Wednesday, December 12, 2018

OpenTrons OT-2 in the house!

YEAAAAAAAH!  Finally here! More details will certainly follow as we (well... @jenkins_conor) writes the software to make this thing:

1) Do BCA assays (whatever those are)
2) Digest things
3) Phosphopeptide enrich things
4) Extract metabolites
5) Dilute extracted metabolites

So far I'm impressed with the thought that went into the setup, the quality of the materials, and Conor said the developer software for it exceeds expectations. The fact it was 1/10 the price of any other one I wanted didn't hurt either...

You don't get a PC with it, but it connects via WiFi and you may need to have some Python guru hanging around to make it do new things it doesn't know how to do. Of course, method development and validation will take some time, but the manufacturer hosts methods developed and posted by other labs. We'll obviously upload ours as we develop them, if the robot is any good.

We're currently using an older Chromebook that is now running Linux, but the thought of getting a Linux tablet for it once we're confident of the performance and reproducibility and sticking it to it with a magnet is seeming like a solid idea.

Tuesday, December 11, 2018

Nonhuman proteins in human milk?!?!??

(Image is from @Confused_Cow, a very strange Twitter account I am now definitely following)

What I actually want to put here is this confusing new JPR paper.... the biological implications that sound like this would be completely and totally biologically impossible (all the proteins we eat get chopped up by that stomach acid stuff, right?), let's look at the amount of work that went into this ---

6 humans donated milk.
This milk was separated into 6 fractions by SDS-PAGE
These SDS-PAGE fractions were analyzed using a passive split flow Agilent nanoLC system in what looks like around 45 minute gradients on a Q Exactive Plus system
MS1 resolution of 35,000
Top10 experiment
MS/MS resolution of 17,500 with a 120ms fill time.

The data processing was done this way --

--I've highlighted the things that I find interesting here, and that I would have done differently.

Wooooooo.....okay.... In one sentence -- This is a really big search matrix searched against masses obtained at medium level resolution, using search tolerances that are strikingly high and searched against a software package version that has been lampooned for it's inability to accurately quantify FDR when matrices get really complex. Yo. If you're Q Exactive Plus is off by 48ppm, get a fire extinguisher. No QE+ that isn't currently on fire or under water is off by that much.

EDIT/Clarification: Proteome Discoverer 1.4 only controls FDR at the PSM level. All later versions of PD were designed to control FDR at multiple levels in order to more accurately control data quality with extremely large numbers of spectra or with high complexity search space.

Sorry, but this seems like an invitation to a list of proteins that are gonna be bad.


If you do your searching of your global and then validate the results with Parallel Reaction Monitoring (PRM) -- and it works -- maybe I should shut up.

Particularly, I guess, if you make heavy isotopic standards as well for your PRMs and this stuff looks legit. And if you go to the supplementary info -- it looks on point --

-I should probably definitely shut up.

In the end -- I'm still confused. WHY ARE THERE DOG PEPTIDES IN HUMAN MILK?!?!?

But the work looks solid.....