Mint Dataset for Human PPI

Question

[beingseema03]

Hi all,

I have downloaded MINT Dataset for Human Protein Protein Interactions. The Link from where I downloaded the dataset is here..-->>> ftp://mint.bio.uniroma2.it/pub/release/txt/current/2012-10-29-mint-human.txt

The dataset is last updated in 2012. Actually my question is that the dataset contains to many fields in the columns of baits and preys as blank entries denoted with dash. That means the baits field is blank and the preys field has the protein name. And also there are some fields with Not protein names but they contain baits and preys fields with numeric entries instead of character names of proteins. And there are some field which are having non-human fields entery and also there are some baits and preys field contain the several protein names for a single column. So do I need to filter out these fields or should I keep these fields?

Answer 1

[Girolamo]

Why do you want to use MINT? any particular reason?

Moreover, Mentha from the same group is updated

[beingseema03]

For my thesis work i have to analyze the PPI databases. Right now i am only focusing on the databases available for Human PPI.
So i chose to analyze IntAct,MINT,HPRD,DIP and Reactome databases. Does this Mentha database contain PPI for humnan..??

Answer 2

[Jean-Karim Heriche]

MINT data are now directly included in the IntAct database.
Regarding how to deal with the data and in particular filtering entries, it all depends on what you want to do.

[beingseema03]

Hi Jean thank you for reply. So that means now MINT and IntAct database are now equal?

And right now I am focusing on only the human database. I have downloaded the species based IntAct database for human.The link from where I have downloaded the dataset is this -> ftp://ftp.ebi.ac.uk/pub/databases/intact/current/psi25/species/

But in the human.zip file there are so many xml files so which particular file do I need to consider for analysis? And also there are some xml files having name as negative.xml. What does this negative.xml file mean exactly?

Kindly help me out.

[Jean-Karim Heriche]

IntAct always (at least for the past few years) had the MINT-curated data. It's just that they used to be first put into the MINT database then imported into IntAct. The process has been streamlined and MINT upload their curation directly into IntAct. If you want/need a comprehensive set of human protein interactions, then I suggest you use iRefIndex. Otherwise, if you need something absolutely up-to-date, you have to assemble the set yourself from the various databases.
The negative.xml files in IntAct contain pairs of proteins that are known NOT to interact. The multiple files represent subsets (by publication ?) of the whole data. However, the entire data is also available in a single file in tab-delimited format here.

[beingseema03]

Hi Jean,
Thanks for you reply. It really helps alot. The link which you provided for whole dataset is the dataset of Human PPI..?? Or it includes every kind of PPI.Because i am just focusing on Human PPI.