Wednesday, October 21, 2020

Need THE guide for today's proteomics for medical collaborators?

 Is this the guide to today's proteomics technology that Carl Sagan would have written? Something that breaks down the crazy stuff we do that is branching into hundreds of distinct directions and makes it as approachable as possible? 

If not, it's as close of an attempt as I've ever seen and it makes me very happy.  100% recommended for sending to that person you think will be cool to work with.  

Monday, October 12, 2020

STAMPS -- Build Metabolomics Assays From the Protein Level!


I'd love to have an argument with you about what "Metabolomics" actually means. To me, the sticker is that whole "-omics" part of the word. Wait. Terrible idea time. 

I'll argue all day that "proteomics" is only just now starting to happen routinely. Here is the argument, though, I'm not sure "metabolomics" is really happening yet. We know that only a small subsection of the metabolites in a cell will stick to reverse phase chromatography, and an even smaller section of those will ionize in one polarity and an even smaller section of those will survive ionization and out of those you'll confidently identify maybe 10% , maybe 30% of them?  If you haven't tried global metabolomics, give it a whirl. Then tell me how you figure out which of the 25 things confidently labeled as "inosine" is inosine. (The second one is probably hypoxanthine, it doesn't survive electrospray very well.)  

What was I typing abou....STAMPS! You can read about it here.

STAMPS dares to ask the question: "Why are you trying to quantify these frustrating small molecules, when you could be quantifying the proteins responsible for making and degrading them?"

And you say: "Because it would take me 3 miserable months to make the targeted assays"

And they said: "Oh. Here you go. We made them all for you." (In mouse, so far, but more is clearly coming)

This thing is super sweet. Go into the database and find the metabolic pathway you're interested in. (They've got 16,000 proteins in mouse available so far) 

Put a checkmark on the proteins that you want, right in the metabolic pathway for the thing you're interested in. 

Check the spectra of your targets, if you're interested. 

And download your assay. Formatted for input into Skyline! 

Is this still sorta targeted protein quan and not proteomics? Maybe. But if you've got the ability to choose from all the things, to me, that counts. I'll be even more pumped when they set up human, but if I was really interested it wouldn't be all that hard to port this with Picky or Phosphopedia or similar. 

Sunday, October 11, 2020

PRiSM -- A thought experiment on Protein rather than Peptide Spectral Matches!

(Image credit: Lucas Vieira for making this and putting it in the open domain!

Proteomics hasn't been around all that long (not real proteomics, anyway) but we have been around long enough to develop a nice cognitive box for us to work within. 

The answers to to these questions are something like:
1) We do the peptide because it is WAY easier. 
2) You can do it. I mean....they did it.... I couldn't do it. 
3) You need some serious math and a lot of firepower. 
4) Pros? It totally works and they can find things that traditional engines can't.
5) Cons? It's hard. Like seriously hard. 3 hours per MS/MS spectrum per computational core hard, but this is a thought experiment, not a practical optimization study. 

We know there are places that our bottom-up search engines just don't work well. Maybe there is an alternative! 

Saturday, October 10, 2020

Do you have a strong Opinion on Proteomics? This special edition wants to hear it!

 I know you have some opinions about what is good and maybe what is bad in proteomics right now. (I'm anything could possibly be bad about proteomics!) If you're tired of saying that opinion over and over again to yourself, and think the whole world should hear it -- now is your chance, cause this special edition in Proteomes is accepting reviews now. 

How much do they want your review? No page charges! (Don't tell anyone at Elfseverer about this. They might literally die.)  

I'm not kidding about it being an opinion piece, special guest editor Matthew Padula described his hope of assembling a "warts and all" compilation of where we are today, so we can take it apart and figure out how to fix it all to get to the next level.

Super cool idea, right? 

Friday, October 9, 2020

Reminder you can control your EasyNLCs from your PC!

 Probably everyone already knows this? Just in case, you can totally control your EasyNLC systems from your computer. It makes remoting in a lot easier. I just confirmed today that it is compatible with the Easy1200 system.

In our test group of mass spectrometrists considered "grumpy" and "deficient in Vitamin D" by their colleagues, we found a 12% reduced incidence in road rage when they were running the horrible "flush air" script on their horrible commutes. 

You can get the installation documents from the great UWPR page here

Thursday, October 8, 2020

Fragment mass prediction for phosphosite localization!

Could we improve on current phosphorylation site localization strategies? This group sure seems to think so and this new approach seems simple enough with all this deep learning peptide stuff happening to work into a lot of new workflows! 

If I get what they're doing, they use the deep learning model for the unmodified peptides and use the phospho and phospho loss (?) shift to score the localization. The plots vs the traditional approach suggest they're onto something solid. 

Wednesday, October 7, 2020

The Carrier Proteome Effect in Single Cell Proteomics!

 I saw this one Twitter a few days ago and I couldn't find it anywhere. I think there might be a preprint, but I'm leaving it here so I don't lose it again.

Direct video link is here. 

Tuesday, October 6, 2020

Match between runs for Reporter Ion Quan?!?!?


Have you ever tried to combine multiple TMT studies? I'm paraphrasing, but Akhilesh Pandey said "two will work, and three is okay, but the amount of loss by your fourth plex makes it not worth it" and I have found that to be 100% true in my hands. Our sampling is stochaistic, which is fun to say, but I'm not sure that I'm using the word correctly. The Venn diagram of the peptides that you're able to fragment and identify in each plex that you add gets progressively less overlapping. In today's instruments that are blindingly fast, we're getting loads more fragment ions but that doesn't necessarily translate to higher percentages of actual identified ones.

Isobaric match between runs (IMBR) allows you to link data from fragmented but not identified peptides back to the identified ones from other plexes. 

I'm legit blown away. There are at least 10 different studies on hard drives sitting around here that this should help with. 

Oh, and they built a new algorithm to normalize without using a pooled channel. 'Cause, you know, this wasn't incredible enough. And these appear to now be built into MaxQuant and Perseus now! 



I was on a conference call the other day and someone said that we needed more tools for combining TMT plexes. If you're not about to start updating your MaxQuant after reading this post, you should 100% check out IRS from Phil Wilmarth. You'll need to use R to do this, but you can pretty much follow anything Phil does and copypasta what he writes directly into R and just run it. 

Wednesday, September 30, 2020

My argument for why MS should be on the the front line for emerging pathogens!

 The good people at Bioanalysis Zone let me write some of my opinions up and edit them for a more mainstream audience than this weird blog normally appeals to. In this installation, I talk about just a few of the shining successes MS labs that were able to get access to SARS-CoV-2 samples have had. I also whine about why there were so relatively few of them. 

I guess I tried to make an argument that based on the ratio of (access to sample valuable contributions) that if we had more access there would clearly be more contributions. Not sure if it carried across the way  I meant, or not. I mean...if you clearly know what you're doing (not talking about me, obviously, but other people)...and you want to help, you shouldn't have to go to Craigslist or LinkedIN to find disease samples during a freaking pandemic, right? That's what legitimate people in legitimate roles have had to do. 

Imagine if we'd had an established front line of defense network of labs (featuring mass spectrometry) that had the job of responding to emerging pathogens, providing testing until the next gen technologies could be scaled up and providing drug activity data from day 1.....  How different would it be today? 

Tuesday, September 29, 2020

MetaMorpheus -- Now with more style (and some cool visualizations)!

I know, I know, not everyone loves Metamorpheus and MSFragger and their every update shouldn't qualify for some kind of a lazy blog post, but they are both getting better all the time. This new update for MM has both a great gain in asthetics as well as some super handy visualization features. 

Ask yourself this -- how many PTMs did you search with your tool of choice today? Did you manually click "static + carbamidomoocowdoacetylatoin" and "dynamic + oxidation" and hit the go-button? Maybe you did the pyro-N-Glu thing? 

Chances are your list looks very similar to what Metamorpheus found in that represents the largest bars on this cool histogram below. 

 I didn't punch in any PTMs at all. I just added my workflows and hit go and I got what you did and all this other stuff!  I was lazier and got more data. Seriously, though, the PTM visualization thing is really cool. 

Monday, September 28, 2020

Toward a proteomics META-DATA standard!


Have you been thrilled to find that there is publicly available data for a project similar to the one you're about to start and had your hopes dashed against sharp rocks because the files uploaded make little sense? 

Have you thought about testing your new data processing pipeline for someone's old data and read 300 pages of supplemental information to discover they never tell you what file is what? 

The thing that is missing from that experiment is MetaData. (Pronounced like "Mee Ta Da Tay; imagine that 90s star wars character with the big ears saying it. It's like that.) 

And a bunch of busy bodies are trying to fix this problem by annotating data previously deposited and by giving us some templates for how we can annotate the things we upload. Strongly recommend you check this out. It'll be great for all of us! 

Sunday, September 27, 2020

Characterisation of protease activity during SARS-CoV-2 infection!

SARS-CoV-2 infected cells.

Drug curves! 


Weird protease activities (viruses are dumb) 

Pretty graphs! 

Publicly available data! 

A corresponding author who just set up his lab a few months ago and already has data? Seriously impressive work all-around. 

Wednesday, September 23, 2020

Is proteomics ready for the clinic? Perspectives on Acute Myeloid Leukemia.

This great new review at IJMS asks the question: "Hey! With all this better proteomics stuff, can you help us with leukemia yet?" 


I dig this as a summary of both the advances in LCMS technology that could feasibly be used within the confines and limitations of a medical environment (double dog dare you to try pitching offline fractionation to a hospital administrator) as well as a summary of the really promising work that has been done in one specific disease.

Table 2 is beautiful and says a ton about the potential of proteomics and our traditional limitations. It says to me: "Here are great studies! Can you imagine what we could do with larger cohorts...?" We'll get there!

Tuesday, September 22, 2020

Complex proteomic patterns in chemotherapy response in breast cancer (113 patient FFPE!)!

I can't follow all the cancer terms in this paper despite hearing lots about HER2 and some of these other things over the years. What I can follow is this is a big cohort multi-omics study that appears to have been done very very well

It represents a tremendous amount of work, starting with 113 FFPE tissues from patients broken down by genetic representation of their tumors and the treatments they received, as well as how they responded to the treatments they received. I think it represents something like 30 individuals, given the pre- and post- treatment samples. On top of this cell lines were also used for both proteomics and metabolomics. 

Super-SILAC was used for the proteomics from the laser microdissected (pretty sure that's how they did it?) samples as well as the cell lines (?) a little fuzzy on the design here without digging in much further and the metabolomics utilized heavy glucose and glutamine. The LCMS for the metabolomics is completely new to me, and I think it deserves exploration. Something called a zipHILIC was used with a low flowrate (100uL/min with AmBiC/ACN gradient and a 59 min gradient!) 

The study keeps going. They use CRISPR and do some mouse work to support their findings and they even use a SeaHorse (not the mythological creature, the Agilent high throughput metabolism thingy). 

This is a really inspiring amount of work with a great story about the value of proline metabolism in chemotherapeutic response. 

690 RAW files can be downloaded at PRIDE here (PXD012000). I haven't found the metabolomics ones yet. This may just be the proteomics!'s somewhere in the range of a freaking TERAbyte of RAW data!!) 

Monday, September 21, 2020

Interpreting peptide fragmentation -- spectral quality overrides software!

This is a FANTASTIC resource for just about anyone! 

Is this peptide identification real? Why or why not? 

This guide is open access, clear, and covers everything! 

100% recommended if you're wondering about that PTM or if you're new to manual interpretation of mass spectra! 

Saturday, September 19, 2020

Glycoproteomics of sparkling wine?

 Need more proof that glycopeptides are involved in everything? 

I suspect that this paper is about the fancy stuff that no one is allowed to called Champagne unless the grapes in it can be directly seen through the bathroom window of one specific bureaucrat in a really dreary looking part of France, rather than the sparkling wine that people make into slushies in Pennsylvania. Just in case you needed more reasons to look forward to ASMS 2020 Philly!!! 

Back to the fancy stuff! 
This team uses SWATH to anayze different sparkling wines fermented in different ways, while keeping in mind desirable characteristics. Turns out there are a load of glycopeptides and the amount of time the wine hangs out on the lees has a lot to do with their distribution.

Interestingly they find a lot of nontryptic cleavages, suggesting that there are some proteases from the grapes or wood or yeast or something at work in the bottles! 

The data processing is done with Byonic, which I didn't know could process DIA data. 

Friday, September 18, 2020

hu.MAP -- Learning machine magic on 15,000 proteomics experiments!

 This new preprint outlines a project that could be termed "ambitious"....

Wooo....ummm.....shit..... I was really going to try and write something clever about this that at least created an illusion that I had some idea what was going on here. Nope. Not going to happen.

These people took an absurd amount of data and tried to come up with a better way of predicting protein-protein interactions. We have great actual protein-protein interaction data like BioPlex, so it makes sense that if your fancy artificial netneuro learny machine doohicky could learn from the sets and match the data from BioPlex, you're probably on the right path, since that's kind of the gold standard and everything. I think they show that in one of the curves. 

After kind of being able to follow along for the length of this paper, it seems like when you go to the hu.MAP 2.0 database that you'll be blown away by the awesome and powerful data at your fingertips... so I punched in a couple of proteins I know really well....and....

....I suspect it's way cooler if you read the instructions. If "RAS interacts with RAS Interactor #1" is what came out of 15,000 RAW files, we might still be a lot closer to...

....than Cyberdyne System (Boston Dynamics) rolling out a Terminator. 

Tuesday, September 15, 2020

MASSIVE.Quant -- Reanalyze, merge, reimagine publicly deposited data!

This is far far too cool for anything I can write to express how cool it is.

This makes it look like we can reanalyze any quantitative proteomic dataset with virtually any tool that we have in our utility belts.....

 You need to check this out. 

YPIC Mini Challenge!


What were you planning to do this weekend? Virtual Zoom based clubbing? Me too! 

Around that? What about a mini challenge set up by our friends at the London Proteomics Discussion Group and the EuPA YPIC group? Information is in the thing at the top.

And if you aren't completely and totally maxed out on COVID-19 proteomics stuff, the LPDG is still going --- Ray Iles is presenting this Friday (presumably an update on the great MALDI-TOF detection of the virus work that posted on MedRXiV just recently) as well as Akhilesh Pandey now of the Mayo Clinic talking about PRM of SARS-CoV-2 peptides. These seminars are this Friday and you can register for them here.

Monday, September 14, 2020

A (modern) beginner's guide to mass spectrometry-based proteomics!

 This new and modern and open access beginner's guide to mass spectrometry proteomics makes me very happy. I've added it at the top of the permanent "Resources of Newbies" page over on the right side of the screen somewhere --->

Some of you ridiculous people are great at mass spectrometry and biology things. I'm not. I can run a mass spec, type fast and...that's....about the end of my life skills. My attempt at a bathroom remodel put me in the hospital with a screwdriver shaped hole in my eyeball (99.8% of my vision back already!). I NEED (safety goggles at all time, apparently) biologists who are patient enough to explain repeatedly what they need the magic mass spec to do. If they understand what the limitations of the devices are and that it really isn't magic? Conversations are 1 million times easier and we can design a real project. This is one of those papers to send to people who have cool sounding biology things that you want to help them with to get those conversations going. 

Monday, September 7, 2020

Synergistic optimization of LC and MS parameters on Tribrid instruments!

Okay -- we all know how to set up the perfect DDA experiment on a Tribrid, right? I sure felt like I had it locked down, but I learned a lot from this incredibly thorough and inciteful new study. 

The goal was to start with one of the most important studies ever written for us proteomics nerds in the lab and see where a Tribrid stands in a similar analysis. How many features are there? How many can we get with this thing? What are we missing and why? 

There is a lot in here, but I'll just go over the highlights. 

1) One of the most interesting findings might be some really suboptimal results using the "top speed" methods. The authors find that even though they predict 50 OT or 60 IT scans should be occurring per cycle, they only obtain 18 OT or 12 IT scans per cycle. Ouch!  I need to check some old data later to see if I'm seeing anything close. 

2)  By analyzing the amount of time needed to get to 50,000 charges in MS/MS, they find the following:

97.4% of the peptides required >20ms of fill time to get there. Totally agree. 
22.4% of acquired MS/MS scans require >150ms to get to that same state (lower abundance)
And....they cite a study I should really already know that suggests that for the ultra-low abundance stuff, 2 seconds is required to get to that target. Two(2) seconds?  Whoa. 

3) Okay -- and this is super cool.
The authors look at the reproduciblity of label free quantification at different flow rates, gradient lengths, and loading. 
High load? Worse quan. Sure, it looks prettier to have tons of signal, but the quan falls off a cliff. What they find is that when the MS1 fill time drops to <1ms you lose some dynamic range in your experiment. 
This makes sense, right? Automatic Gain Control (AGC) is trying to keep the C-trap and Orbitrap from overloading by capping your fill time. Is it more accurate when it has 10ms to compensate or 0.1ms? I'm going to guess it gets better at the higher numbers.
They find the best quan on their 20cm 1.8um columns at <1ug and do not recommend going over 2ug regardless of the gradient length on the column. 

Is this the most exciting study of all time? Maybe not, but the findings here are definitely worth thinking about before you set up that next global study on your Tribrid. 

Thursday, September 3, 2020

COVID-19 Multi-Omics Viewer!

Got some COVID-19 fatigue? Me too!  So I'll keep this short, but it's way too cool to not say something about it. The preprint is at medRxiV here. 


Don't feel like reading (or accidentally stabbed yourself in the eye with a screwdriver and that eye is predicted to work at 100% again in about 2 weeks, and reading suuuuuucks in the meantime? []?) 

I always think its got something to do with integrin something or other. is that the only protein family I know by name? Maybe? But that great chart at the top suggests that in 129 patients(!!) maybe Integrin beta 3 does differ between severity types?  Just an awesome resource!   

More great stuff out of Wisconsin, which is in the news right now because the people in power are some of the worst human filth to ever walk the surface of this planet. However, it's really interesting that it isn't really reflective of the general population of the state. The state is heavily heavily "gerrymandered" meaning that lines for specific districts are designed to purposely bias the system toward the candidates on the right. How biased? Really amazingly biased! 

Left is popular vote and right is seats won! Amazing, right?  It's a systematic problem in the US., particularly in states that are very important for the presidential race that is coming up in just a few months.... For real, not all of us here are morons. We are, however, a country of incredibly lazy people and we've let really evil people twist our system into what it has become today.

...what was I supposed to be doing...? Oh yeah! Tracefinder stuff! 

Pushing the limits of a Q Exactive "Classic"!

 I think the preprint of this paper might have made the blog a while back, but even if it did (or some iteration -- preprints do often evolve) it's worth revisiting. And...if I read it and meant to write something and forgot? Even better! 

This is the study ASAP at JPR! 

How far can you reeeeeeeeeeeeeeeeeeeeeeaaaaaaaaaaaaaaaaaaaaaallly push a Q Exactive "Classic" system if you pulled out the stops? 

Like you: 

Really really really optimized your DIA windows?

You used some fancy micropillar chromatography for long gradient single shot? 

You used even fancier predicted libraries? 

Could your "Classic" system come back with 8,000 protein confident protein IDs in a single run? You'll have to read it to find out, I guess.