What you're asking here is essentially a crash-course in being a DBA. I'll try to explain this by sequentially explaining how the Oracle Database software comes to life at startup.
An "instance" is the Oracle software running in memory, before it touches any files. It goes away when the power is switched off.
init<DB>.ora is the instance parameter file. (This is a human-readable PFILE, which is one of two kinds of parameter file. The other is a binary SPFILE.) This is what defines the parameters to start the Oracle Database software in memory, such as how much memory to use. Yours may not look like theirs, and it doesn't have to. The most important parameters here are the database name and the location(s) of the controlfile(s).
After the instance starts and allocates memory, it then "mounts" a controlfile. This defines the other files that are available, the state/version level of transactions in the system, and other extremely high-level metadata. Note that we're still not up to the part that considers the user data in the database. Like an SPFILE, a controlfile is stored in a binary format, but also like a PFILE, the controlfile can be backed up "to trace" to generate a human-readable means of recreating it. Note that only the binary version can actually be used.
The parameter file defines the instance, and the controlfile defines the database.
A "database" is the collection of files on the disk. This is where the business data and the logical metadata is stored. After the instance mounts the controlfile, it reads the controlfile and "opens" the database. If it's missing any data files--or they fail consistency checks--it won't open.
The challenging part of answering this properly is that both the instance parameter file and the controlfile(s) will likely need to be recreated on your end if you're going to have any luck accomplishing what you'd like to do.
This is not unlike duplicating a production database to a test environment, detailed here: http://www.dba-oracle.com/oracle_tips_db_copy.htm