4 ms·

Proteomics with OMSSA

29 points by davikr 3y ago

hijodelsol 3y ago
While interesting, I think that the relevance of MassSpec based approaches will decline over the next years. Single cell genomics already had to learn the hard lesson that pure gene quantification and big data approaches without spatial resolution or deep understanding of the biology can only get you so far. Billions have been spent on cell atlases that just deal with biology as individual cells without much to show for in terms of changes in clinical practice. Spatial omics is now finally spreading and spatial organization and interactions are being taken into account again, but the common analysis pipeline still just focus on gene level expression at the cellular level. Subcellular and extracellular information is often completely lost and gene levels are highly variant and only somewhat correlate with the phosphoproteome, the actual effect layer of biology. Mass spec approaches share many of the same limitations, even though they deal with protein. Spatial information is blurry at best or lost at worst and equipment is expensive and requires specialized training. Imo, the most interesting advances in the next years will come from low cost high resolution spatial proteomics with high target counts that integrate with biological modeling of processes.
- getoffmycase 3y ago
  I disagree. I think that proteomics is useful without spatial information because it captures a snapshot of cell state, or tissue state rather, of signaling networks via PTMs along with abundance measurements. If you don’t have spatial orientation, you at least have temporal information. I think increasing the yield of MS2 scans to PSMs specifically by dealing with spectra containing PTMs and chimeric spectra will further enable deeper understanding of cell signaling. Additionally, the targeted analysis of specific sub-proteomes using real time search, using GoDig from the Gygi lab for example, also seems very promising. Plus, there’s large industry efforts using proteomics as a drug screening tool, an application that doesn’t require spatial resolution of anything. Specifically groups are looking for protein expression knockdown, but it’s not too far a stretch to look for pathway perturbations using real time search and careful controls.
pfd1986 3y ago
Does anyone know a good general audience book on proteomics? It seems to be a very interesting field with a bright future but I can never find the time to dig deep
- deng 3y ago
  It really depends on your prior knowledge and what aspect of proteomics you are most interested in. A good starting point is here: https://proteomicsnews.blogspot.com/p/resources-for-newbies.html https://proteomicsnews.blogspot.com/p/resources-for-newbies.... Also, if you haven't already, maybe take a look at transcriptomics first, as there's more introductory material on this topic. For instance, there's an excellent online course at https://diytranscriptomics.com/ https://diytranscriptomics.com/.
- pfd1986 3y ago
  Actually, to answer my own question, this seems like a good intro article: https://erictopol.substack.com/p/spatial-biology-is-lighting-it-up https://erictopol.substack.com/p/spatial-biology-is-lighting...
- jszymborski 3y ago
  IMHO, most fields in biology lack good on ramps when compared to CS. I think part of the reason is that it's way easier to get someone to install Python than to get access to a mass spectrometer. A big factor in me studying Biochemistry in undergrad was that it was way easier to self-teach myself the parts of CS that interested me than it was for Biochemistry, in my estimation. I'm sorry I don't have a better answer for you, I hope someone else here does!
  - maegul 3y ago
    My experience in grad school was similar. IMO it decelerates the shift from undergrad to grad work, as you have to overcome the knowledge gulf by immersing yourself in the essentially confusing sea of primary literature. Arguably unnecessary IMO … good material at various levels of expertise makes a lot of sense.
- 3y ago
pugworthy 3y ago
For the curious... OMSSA = Open Mass Spectrometry Search Algorithm It seems like a lot of 404's on the various NIH links for it, but I did find the FTP download is still at https://pubchem.ncbi.nlm.nih.gov/omssa/ https://pubchem.ncbi.nlm.nih.gov/omssa/ Here's someone's command line option documentation for it -> https://sivome.github.io/proteomics/2019/03/02/Proteomics-with-OMSSA.html https://sivome.github.io/proteomics/2019/03/02/Proteomics-wi...

29 points by davikr 3y ago

While interesting, I think that the relevance of MassSpec based approaches will decline over the next years. Single cell genomics already had to learn the hard lesson that pure gene quantification and big data approaches without spatial resolution or deep understanding of the biology can only get you so far. Billions have been spent on cell atlases that just deal with biology as individual cells without much to show for in terms of changes in clinical practice. Spatial omics is now finally spreading and spatial organization and interactions are being taken into account again, but the common analysis pipeline still just focus on gene level expression at the cellular level. Subcellular and extracellular information is often completely lost and gene levels are highly variant and only somewhat correlate with the phosphoproteome, the actual effect layer of biology. Mass spec approaches share many of the same limitations, even though they deal with protein. Spatial information is blurry at best or lost at worst and equipment is expensive and requires specialized training. Imo, the most interesting advances in the next years will come from low cost high resolution spatial proteomics with high target counts that integrate with biological modeling of processes.

getoffmycase 3y ago
I disagree. I think that proteomics is useful without spatial information because it captures a snapshot of cell state, or tissue state rather, of signaling networks via PTMs along with abundance measurements. If you don’t have spatial orientation, you at least have temporal information. I think increasing the yield of MS2 scans to PSMs specifically by dealing with spectra containing PTMs and chimeric spectra will further enable deeper understanding of cell signaling. Additionally, the targeted analysis of specific sub-proteomes using real time search, using GoDig from the Gygi lab for example, also seems very promising. Plus, there’s large industry efforts using proteomics as a drug screening tool, an application that doesn’t require spatial resolution of anything. Specifically groups are looking for protein expression knockdown, but it’s not too far a stretch to look for pathway perturbations using real time search and careful controls.

pfd1986 3y ago

Does anyone know a good general audience book on proteomics? It seems to be a very interesting field with a bright future but I can never find the time to dig deep

deng 3y ago
It really depends on your prior knowledge and what aspect of proteomics you are most interested in. A good starting point is here: https://proteomicsnews.blogspot.com/p/resources-for-newbies.html https://proteomicsnews.blogspot.com/p/resources-for-newbies.... Also, if you haven't already, maybe take a look at transcriptomics first, as there's more introductory material on this topic. For instance, there's an excellent online course at https://diytranscriptomics.com/ https://diytranscriptomics.com/.
pfd1986 3y ago
Actually, to answer my own question, this seems like a good intro article: https://erictopol.substack.com/p/spatial-biology-is-lighting-it-up https://erictopol.substack.com/p/spatial-biology-is-lighting...
jszymborski 3y ago
IMHO, most fields in biology lack good on ramps when compared to CS. I think part of the reason is that it's way easier to get someone to install Python than to get access to a mass spectrometer. A big factor in me studying Biochemistry in undergrad was that it was way easier to self-teach myself the parts of CS that interested me than it was for Biochemistry, in my estimation. I'm sorry I don't have a better answer for you, I hope someone else here does!
- maegul 3y ago
  My experience in grad school was similar. IMO it decelerates the shift from undergrad to grad work, as you have to overcome the knowledge gulf by immersing yourself in the essentially confusing sea of primary literature. Arguably unnecessary IMO … good material at various levels of expertise makes a lot of sense.
3y ago

pugworthy 3y ago

For the curious... OMSSA = Open Mass Spectrometry Search Algorithm It seems like a lot of 404's on the various NIH links for it, but I did find the FTP download is still at https://pubchem.ncbi.nlm.nih.gov/omssa/ https://pubchem.ncbi.nlm.nih.gov/omssa/ Here's someone's command line option documentation for it -> https://sivome.github.io/proteomics/2019/03/02/Proteomics-with-OMSSA.html https://sivome.github.io/proteomics/2019/03/02/Proteomics-wi...