Dear Ajay Sarawagi,

there are different ways to increase the number of protein identifications, but they are all leading to a higher number of samples to be processed in the LC-MS.

First I would try to separate the mouse brain in severeal areas. I have done this with rat and with mouse brain for a lot of experiments and it will improve very much the level of the detection of changes between treated and untreated animals. Subfractionation of the tissue will also be benefitial because most processes were happening at membranes more than in the cytosol which has approx. 70 to 80% of the total protein content in mouse brain. You can look for our publications how we have done a subfractionation for rat brain to get different fractions such as nuclei, synaptosomes, mitochaondria, myelin, microsomes and cytosol. Another very well working method that I am using is the prefractionation with isoelectric focusing of the tryptic peptides before doing lc-MS. By dividing the single sample (with approx. 5000 identified proteins) into 6 fractions with different pI's the number of identified proteins can be increased up to 10000 (for method description look at my book chapter:

IPG Strip-Based Peptide Fractionation for Shotgun Proteomics May 2014 · Methods in molecular biology (Clifton, N.J.) 1156:67-77 DOI: · 10.1007/978-1-4939-0685-7_5

You can also do a 2D HPLC separation to increase the number of identified proteins by doing a prefractionation with high pH c18 RP- chromatography before going on with the standard low pH c18 RP- chromatography before MS.

But as I mentioned before, in evrey cas you will need to increase your number of LC-MS measurements.If you do not have free access to a well adjusted and supervised MS system you need to check for the capacity that you are able to run.

Good luck!

Murat

Dear Ajay Sarawagi ,

the simplest answer would be: just follow protocol which was used in articles you are referring to. If it does not work as it supposed to, try to find a problem. I would guess that your mass-spec is a problem. Have you run any of cells extracts (HeLa or K562) to check a performance of your machine ? What kind of mass spec are you using ? You mentioned Q Exactive but there is a huge difference between Q-Exactive (normal) or Q-Exactive HFX. You have to also take it to the account. I identify usually 4-5k proteins from a cells exctract using Pharma Fludics columns in a single run.

As it was suggested by Murat Eravci I would also perform further fractionation. If you would like to stick to gel-based method I would try with big gels and run 300-400 ug of protein extract. Fractionate it to 24-30 fractions and follow my protocol:

https://www.nature.com/articles/s41598-018-26639-3

where you can prepare samples in very effcient manner.

I guess using SEC for protein fractionation would be also helpful.

Other suggestions you will find in Murat Eravci post.

A longer column (25 or 50 cm) would help boost the number of IDs..

also

IDs can change a lot based on search algorithms used and how the peptide results are filtered and used to generate protein hits. You might download a dataset that has high reported protein coverage and run it through your data analysis pipeline for comparison.