A beginner question about eulerian path algorithms

Question

[Abhishek]

Hello all,

I am working on an assignment in which I am to design and implement an eulerian path algorithm for genome assembly using kmers (a la velvet). My question is this: How does the algorithm form kmers from the intersections of reads? I assume it searches for overlap between reads and then forms kmers from that, but what if there is only an overlap of 1 or 2 bases? Wouldn't there be many reads which would then potentially overlap each other?

I apologize if this question is too basic for this forum.