How many transcripts does it take to reconstruct the Splice Graph?

In recent years the phenomena of alternative splicing has emerged as functionally very important and occurring with high frequency. Alternative splicing is typically detected by sampling a series of transcripts that have been made by copying different configurations of exons. The alternative splicing graph (ASG) is then constructed so it minimally can explain these transcripts. The question here naturally arises, how well does this minimal alternative splicing graph correspond to the true alternative splicing graph as a function of trascript sample size.