Lee Cantrell
@leecantrell.bsky.social
Biological Mass Spectrometry
Proteomics at Scale
Scientist at Seer
Opinions independent of employer
Proteomics at Scale
Scientist at Seer
Opinions independent of employer
To clarify, (not speaking for employer), this is the Seer XT nanoparticle product. Customers like Chiara Guerrera at Necker Proteomics have independently evaluated multiple technologies speaking to this. www.linkedin.com/posts/chiara...
Mining the plasma proteome: Evaluation of enrichment methods for depth and reproducibility | Chiara Guerrera | 15 comments
Our paper "Mining the plasma proteome: Evaluation of enrichment methods for depth and reproducibility" is finally out in the Journal of Proteomics! 🎉
This journey started when we were entrusted with ...
www.linkedin.com
October 19, 2025 at 12:17 PM
To clarify, (not speaking for employer), this is the Seer XT nanoparticle product. Customers like Chiara Guerrera at Necker Proteomics have independently evaluated multiple technologies speaking to this. www.linkedin.com/posts/chiara...
Clinical protocols are very tricky to standardize across sites.
September 14, 2025 at 12:43 PM
Clinical protocols are very tricky to standardize across sites.
Yeah just run a sample every 2 minutes for 5 years with no overhead or down time and a 2M instrument costs 1.5 per sample…
Pretend 15 min with 75% up time and it’s 15/sample. Adjust cost relative to purchase price.
Pretend 15 min with 75% up time and it’s 15/sample. Adjust cost relative to purchase price.
August 3, 2025 at 2:42 PM
Yeah just run a sample every 2 minutes for 5 years with no overhead or down time and a 2M instrument costs 1.5 per sample…
Pretend 15 min with 75% up time and it’s 15/sample. Adjust cost relative to purchase price.
Pretend 15 min with 75% up time and it’s 15/sample. Adjust cost relative to purchase price.
Isomerization? Especially in deamidation prone peptides with NQ. This does split peaks in LC. Eye lens work has seen this for decades. Doesn’t mean deamidation necessarily. Can generate 4plet or more if including modification. Tris in prep?
July 8, 2025 at 8:49 PM
Isomerization? Especially in deamidation prone peptides with NQ. This does split peaks in LC. Eye lens work has seen this for decades. Doesn’t mean deamidation necessarily. Can generate 4plet or more if including modification. Tris in prep?
Or on high mass? I haven’t seen much for >100ng from the 8600.
June 28, 2025 at 2:41 PM
Or on high mass? I haven’t seen much for >100ng from the 8600.
Not that I’m aware of - though I’d also be nervous about charge capacity on transmission through the tims device. I’d rather see the 7600 or 8600.
June 28, 2025 at 1:17 PM
Not that I’m aware of - though I’d also be nervous about charge capacity on transmission through the tims device. I’d rather see the 7600 or 8600.
It’s unclear that SLIMS has the reproducibility or resolution to achieve this. Marketing figures aside, peptides don’t always make nice Gaussians. Co-resolving charge states likely isn’t an issue.
What I want to see is an honest effort on DIA vs PAMAF with a good DIA TOF instrument.
What I want to see is an honest effort on DIA vs PAMAF with a good DIA TOF instrument.
June 28, 2025 at 12:32 PM
It’s unclear that SLIMS has the reproducibility or resolution to achieve this. Marketing figures aside, peptides don’t always make nice Gaussians. Co-resolving charge states likely isn’t an issue.
What I want to see is an honest effort on DIA vs PAMAF with a good DIA TOF instrument.
What I want to see is an honest effort on DIA vs PAMAF with a good DIA TOF instrument.
Rationale seems to be more ions = better data. Astral transmits ~1/200th of ions within a target search space at a time (less the overhead and MS1 times).
A 400ms SLIMS separation could reasonably replace half the selectivity of the quad and 4x throughput of qTOF.
A 400ms SLIMS separation could reasonably replace half the selectivity of the quad and 4x throughput of qTOF.
June 28, 2025 at 12:30 PM
Rationale seems to be more ions = better data. Astral transmits ~1/200th of ions within a target search space at a time (less the overhead and MS1 times).
A 400ms SLIMS separation could reasonably replace half the selectivity of the quad and 4x throughput of qTOF.
A 400ms SLIMS separation could reasonably replace half the selectivity of the quad and 4x throughput of qTOF.
When I check in every month or so, it’s dry. Much drier than bsky
June 15, 2025 at 1:38 PM
When I check in every month or so, it’s dry. Much drier than bsky
For the right biological question, low IDs may even be preferential if they are subcomponent oriented. I'd argue that any method has its biases, both towards abundance and structural subcomponent.
April 10, 2025 at 3:05 PM
For the right biological question, low IDs may even be preferential if they are subcomponent oriented. I'd argue that any method has its biases, both towards abundance and structural subcomponent.
However, the readout really shows that the sample was not rich in the target subpopulation for that method.
Toolset comparisons should really focus on the same plasma preparation protocol. Ideally there would be robust discussion to the tradeoffs of protocol to implemented toolset comparisons.
Toolset comparisons should really focus on the same plasma preparation protocol. Ideally there would be robust discussion to the tradeoffs of protocol to implemented toolset comparisons.
April 10, 2025 at 3:00 PM
However, the readout really shows that the sample was not rich in the target subpopulation for that method.
Toolset comparisons should really focus on the same plasma preparation protocol. Ideally there would be robust discussion to the tradeoffs of protocol to implemented toolset comparisons.
Toolset comparisons should really focus on the same plasma preparation protocol. Ideally there would be robust discussion to the tradeoffs of protocol to implemented toolset comparisons.
Implementation of any toolset should be done consistently within a project, but comparing study a to study b with different clinical protocols is a bit challenging.
Plenty of techniques now do well for EV proteins. But a double spun sample will typically be low EV. Readout is bad IDs.
Plenty of techniques now do well for EV proteins. But a double spun sample will typically be low EV. Readout is bad IDs.
April 10, 2025 at 2:58 PM
Implementation of any toolset should be done consistently within a project, but comparing study a to study b with different clinical protocols is a bit challenging.
Plenty of techniques now do well for EV proteins. But a double spun sample will typically be low EV. Readout is bad IDs.
Plenty of techniques now do well for EV proteins. But a double spun sample will typically be low EV. Readout is bad IDs.
There's huge dependency on the sample and its consistency in preparation for observables... Double spun plasma is very distinct from single-spun. Storage conditions and a myriad of other variables also impact readouts.
Plasma proteomics is ultimately a tool for clinical sciences to implement.
Plasma proteomics is ultimately a tool for clinical sciences to implement.
April 10, 2025 at 2:56 PM
There's huge dependency on the sample and its consistency in preparation for observables... Double spun plasma is very distinct from single-spun. Storage conditions and a myriad of other variables also impact readouts.
Plasma proteomics is ultimately a tool for clinical sciences to implement.
Plasma proteomics is ultimately a tool for clinical sciences to implement.
Not fully following. But you’ll have MS data files and search result files depending on parameters. There are wrong ways to search, but not a single right way. Upload ideally includes both data and search files.
A 500 sample study can easily exceed TB data size on most any MS.
A 500 sample study can easily exceed TB data size on most any MS.
April 7, 2025 at 12:07 AM
Not fully following. But you’ll have MS data files and search result files depending on parameters. There are wrong ways to search, but not a single right way. Upload ideally includes both data and search files.
A 500 sample study can easily exceed TB data size on most any MS.
A 500 sample study can easily exceed TB data size on most any MS.
Loading 1k astral files searched by Diann in a R interface is relatively painful in my opinion. There’s probably more necessary innovation to search, process and share data from mega cohorts that are now arriving.
April 6, 2025 at 12:36 PM
Loading 1k astral files searched by Diann in a R interface is relatively painful in my opinion. There’s probably more necessary innovation to search, process and share data from mega cohorts that are now arriving.
Unfortunately, MS data is massive. Many observations in a vector multiplexed by LC. New instruments have been “worse” about this.
OLink is quite smart with data presentation being a neat and contained parquet.
OLink is quite smart with data presentation being a neat and contained parquet.
April 6, 2025 at 12:32 PM
Unfortunately, MS data is massive. Many observations in a vector multiplexed by LC. New instruments have been “worse” about this.
OLink is quite smart with data presentation being a neat and contained parquet.
OLink is quite smart with data presentation being a neat and contained parquet.
I’m very pro data sharing. I’ve benefited a number of times from it and been limited by it a number of times. Even available upon request or application data may as well be not shared.
Maybe a challenge is upload. I doubt MOMI 46k astral bio samples will be shared. Well over 1 PB…
Maybe a challenge is upload. I doubt MOMI 46k astral bio samples will be shared. Well over 1 PB…
April 6, 2025 at 12:29 PM
I’m very pro data sharing. I’ve benefited a number of times from it and been limited by it a number of times. Even available upon request or application data may as well be not shared.
Maybe a challenge is upload. I doubt MOMI 46k astral bio samples will be shared. Well over 1 PB…
Maybe a challenge is upload. I doubt MOMI 46k astral bio samples will be shared. Well over 1 PB…
Disagree. There can be legal or IP reasons to not share. Probably why a lot of expensive OLink data wasn’t shared, before. IP can be mined by anyone once out there.
Also human data sharing issues etc.
I wish at least control files would be shared at minimum to validate functionality, though.
Also human data sharing issues etc.
I wish at least control files would be shared at minimum to validate functionality, though.
April 6, 2025 at 12:27 PM
Disagree. There can be legal or IP reasons to not share. Probably why a lot of expensive OLink data wasn’t shared, before. IP can be mined by anyone once out there.
Also human data sharing issues etc.
I wish at least control files would be shared at minimum to validate functionality, though.
Also human data sharing issues etc.
I wish at least control files would be shared at minimum to validate functionality, though.
“chemical and physical biology; structural biology track”
Realistically I was housed in biochemistry and chemistry. Vanderbilt fortunately has two dedicated MS courses, one specifically in proteomics. Another class for using SIMION with John McLean too.
Realistically I was housed in biochemistry and chemistry. Vanderbilt fortunately has two dedicated MS courses, one specifically in proteomics. Another class for using SIMION with John McLean too.
March 23, 2025 at 10:50 PM
“chemical and physical biology; structural biology track”
Realistically I was housed in biochemistry and chemistry. Vanderbilt fortunately has two dedicated MS courses, one specifically in proteomics. Another class for using SIMION with John McLean too.
Realistically I was housed in biochemistry and chemistry. Vanderbilt fortunately has two dedicated MS courses, one specifically in proteomics. Another class for using SIMION with John McLean too.
In grad school - 2018
March 21, 2025 at 12:58 PM
In grad school - 2018