Secretomics Database Query

Thumb default avatar
May 12, 2017

Hi everybody,

As bit of background; I work in a microbiology with C. difficile. I use a gut model to simulate the conditions of the colon, it is inoculated with a human slurry from faeces to populate the model with gut organisms. We can study the populations overtime in a simulated C. difficile infection.

I have been using a proteomics method whereby I can isolate proteins produced from the gut model organisms. These proteins are then passed through LC-MS/MS.

The problem I have is that this environment is very complex and comprises of many different microbial species. There is therefore a great variety of proteins produced, much of the work I have seen has been in single species proteomics.

Is anybody aware of any multispecies bacterial secretomics/proteomics databases I could use to identify my sequences? The proteomics centre that I work with are familiar with identification of human proteins but have not worked with bacterial proteins before. Any suggestions would be very welcome.


Thumb default avatar
mdfenko 10 months ago

are you trying to identify the individual proteins or the organism which secreted them?

the databases available for protein identification by lc-ms/ms should be able to identify the proteins.

you may have to isolate the individual organisms to determine which one secretes which protein(s).

Thumb default avatar
dsp94 10 months ago

Thank you for your reply.

So far the .fasta database used for the main search with MaxQuant was constructed using the available EMBL protein databases from a limited number of bacterial species (who comprised the majority of proteins secreted in my samples).

The data I have been receiving from my proteomics colleagues has been protein identities, and I have been putting those identities through a BLAST search to clarify which organism(s) produce them. Unfortunately, due to the nature of the model I am using, it would impossible to separate the organisms from one another.

The danger with this approach is that even though many proteins are universal in bacteria and are therefore represented in my results, some may be species specific. Due to the fact our colleague is not a microbiologist either, it is likely that the constructed database might not be most appropriate.

In essence I am curious as to what the best approach is to analyse the sequences and which databases are going to be the most appropriate.