Yaamini’s Notebook: SRM Assay Day 1

It’s SRM assay time!

Mass Spectrometer Set-Up

For the SRM assay, we’re using a Thermo TSQ Vantage machine since it is better at SRM detection. To use the machine, we followed a protocol similar to our DIA set-up. Emma attached a previously-prepared 3 cm trap to the mass spectrometer. I set Solvent A to 5% and the flow to 0.2 µL/min using the “trapping” method. I increased flow to 0.5 µL/min after ensuring that the pressure was holding and there were no leaks.

Emma added the first column, but the pressure was low (around 94 psi). She checked the column and found it empty! She then added a new column. I set Solvent B to 5% and changed flow to 0.2 µL/min under the “analytical” method. After a few minutes, the system pressure crashed due to a leak in the trap. This began a series of trial-and-error solutions to ensure the system was running smoothly:

  • Reattached the trap after seeing the leak
  • Pressure crashed again
    • The column itself was long, so it’s possible the pressure inside the column was too high
    • Emma cut the column to 30 cm and tried again
  • Presure crashed: leak at the base of the column
    • Trimmed column once more, replaced t-junction and tried again
  • Pressure reached ~4800 psi, which was too high. The system crashed again
    • Detattched analytical column
    • Added a new trap
      • Broke new trap, added another trap which was 5 cm instead
      • Set method back to “trapping” with Solvent A at 5% and flow at 0.2 µL/min
      • Increased flow to 0.5 µL/min (about 242 psi max)
    • Attached analytical column
      • Flow at 0.2 µL/min, Solvent B 5%, analytical method
  • Pressure crashed due to a leak in the trap
    • Removed column and added a new one
      • Kept old column just in case
    • New column produced a solvent droplet, so I set a ten minute timer
      • Pressure reached ~2780 psi max
      • Switched flow to 0.3 µL/min and 50% Solvent B
  • Trap leaked
    • Trimmed off trap where leak was
    • Reattached trap
    • Tried again at 0.3 µL/min flow

At this point, the column and trap combination finally worked! Once I saw a solvent droplet form at the tip of the column, I set a ten minute timer. Emma took care of resetting the flow and solvent ratios while I moved onto the next step of preparation.

PRTC Addition

I completely forgot about the PRTC addition step before running samples until Emma mentioned it on Friday. We didn’t have any PRTC in the lab, so I borrowed 4-20 µL aliquots from Emma’s lab. I labelled new centrifuge tubes for each of my samples and added 9.4 µL of the final solvent (3% Acetonitrile + 0.1% Formic Acid) to each tube. I started to add 1.9 µL of PRTC to the sample tubes when I realized I probably wouldn’t have enough PRTC for all of my samples. Therefore, I just added it to the samples I wanted to run on the first plate (see plate arrangement below). I then added 7.5 µL of sample to each tube. I vortexed the centrifuge tubes down and then added 15 µL of the solution to each sample glass mass spectrometer vial. For some reason I wasn’t successful at generating 15 µL of mass spectrometery-ready solution for samples 21, 49, 51 and 71, so I had to use another 7.5 µL of raw peptide sample to create a new solution. I now have 25 samples raedy for analysis!

Emma also provided me with 10 µL of a quality control standard. I added 30 µL of Final Solution to this vial and pipetted it into a QC mass spectrometry vial. Emma took 100 µL of the Final Solution and added it into a different vial to use as a blank.

Transition reduction

Emma realized that the 150 transition maximum she provided needs to also include PRTC transitions! We edited down the transitions together. My final transition list can be found here. I also edited my Skyline document and transition selection Jupyter notebook to reflect these changes.

Loading Samples

Because I only had enough PRTC to prepare half of my samples, I split my 50 samples between the two plates in the mass spectrometer. Ideally, I would run the first injection of all 50 samples first, then run the second injection.

Table 1. Plate 1 layout for mass spectrometery. These samples had PRTC added to them and were ready for SRM analysis.

Plate 1 1 2 3 4 5
B O49 O52 O102 O01 O122
C O17 O14 O71 O145 O128
D O39 O113 O12 O99 O103
E O56 O10 O22 O118 O06
F O08 O04 O106 O21 O51

Table 2. Plate 2 layout for mass spectrometry. These samples will be prepared at a later date.

Plate 2 1 2 3 4 5
B O32 O60 O101 O91 O100
C O137 O96 O46 O90 O147
D O30 O31 O131 O35 O24
E O43 O40 O26 O78 O124
F O64 O140 O66 O121 OBLNK2

I had to run to Montlake for some security paperwork, so Emma prepared the queue and started my sample analysis. I should be done with all of my injections around July 18!

from the responsible grad student http://ift.tt/2t6nXgC

Yaamini’s Notebook: Selecting SRM Targets Part 7


After looking at my work from yesterday, Steven suggested that I hit 150 transitions and max out our capacity. However, I’ve already pored over both the short and long protein lists. He offered to help me search for proteins in the full Skyline output. I merged my full Skyline output with annotations to get this list. I evaluated the following proteins that Steven identified as interesting:

screen shot 2017-07-12 at 11 00 12 pm

screen shot 2017-07-12 at 11 00 25 pm

Most of the peaks had interference or missing data, but there were a handful I added to my protein list.

I then evaluated all of my proteins one last time:

screen shot 2017-07-12 at 11 01 59 pm

And I finally had a list of 146 targets from 16 proteins!

screen shot 2017-07-12 at 11 09 31 pm

Aside from some last-minute lab preparation, I’m all set for starting my mass spectrometry runs tomorrow! cue happy dance

from the responsible grad student http://ift.tt/2sSP7Dt

Yaamini’s Notebook: Selecting SRM Targets Part 6

Those peaks are running interference

Emma’s feedback on my preliminary protein targets showed me that I did not peak the best peaks possible for some proteins.

Just as I thought, having one peptide for a protein will not be enough to confidently say that peptide is truly from that protein. Therefore, I will need to delete Extracellular superoxide dismutase from my protein list. After looking at my transitions, Emma noticed that some of my lower abundance transitions were not the greatest.

For example, the transition highlighed in red below has a much lower intensity than the others:

screen shot 2017-07-08 at 2 31 17 pm

And the fifth transition in this peptide isn’t even visible:

screen shot 2017-07-08 at 2 30 09 pm

However, Emma also said that SRM is good at detecting low abundance peptides. It is up to me to decide whether or not I want them. She also pointed out that I had a few peaks that had significance intereference and were pretty sloppy:

screen shot 2017-07-08 at 2 31 44 pm

Peaks like the one above were deleted immediately. I documented my target refinement process in this Excel document. In the end, I had about 135 transitions! I sent my list over to Emma and Steven for feedback one last time:

screen shot 2017-07-12 at 10 43 40 pm

Revised Skyline document

from the responsible grad student http://ift.tt/2sSIMYK

Yaamini’s Notebook: Selecting SRM Targets Part 5

I have draft transitions!

Emma wasn’t kidding when she said sorting through proteins on Skyline takes a while. Even so, I have preliminary target transitions! My steps are laid out in a Jupyter notebook.

To come up with the transition list I sent Emma and Steven, I made a duplicate Skyline document to edit. I followed the rough instructions in Emma’s Skyline tutorial slides to delete bad quality proteins, peptides and transitions.

Deletion criteria:

  • Delete a protein if it only has one associated peptide
  • Delete a peptide if
    • There are less than three transitions
    • There is too much peak intereference, so the peak isn’t clearly defined (i.e. a sloppy peak)
    • There is missing data for samples
  • Delete a transition if
    • It is a precursor ion
    • It has low abundance (want to keep the three most abundant transitions)
    • It is noisy

For example, the least abundant transition in this peak (highlighted in red) is noisy…


…so I deleted it.


While the above photos show the same peptide and transitions, the sample number is different, leading to this peak! For this reason, I needed to use the deletion critera for all samples. If there were more than two peaks between samples, I deleted the peptide.

I first sorted through the proteins Steven marked as “interesting” in my shortlist. Some of those interesting proteins had poor quality peaks, so I looked for proteins with similar annotations to examine instead. Overall, I narrowed 9000+ proteins down to 23 proteins, with 64 peptides and 228 transitions. I was stuck as to how to narrow it down even further, so I figured I would have Steven and Emma help out. Tomorrow, I’ll use their feedback to narrow down my list down to 100-150 transitions.

from the responsible grad student http://ift.tt/2tON7xl