why is paired-end alignment support so important

[found]

Is it because that if a read includes a paired-end, it is not easily to be found out by the current software? but why?

[westerman]

Is it because that if a read includes a paired-end, it is not easily to be found out by the current software? but why?

I am not sure if your question makes much sense.

(A) Yes, of course pair end alignment support is important. One can get more the double the information from paired ends -- e.g., the reads themselves plus the distance between them. Chaison & Pavner have an interesting paper in Genome Research commenting on the importance of paired ends.

(B) While I suspect some alignment programs do not have paired end support it is hardly missing from many packages. It has been around The Roche-provided 'newbler' program does pair-end assembly. The ABI-provided 'corona' program does paired-end matching to a reference. That latter is not assembly per se but then the ABI SOLiD is not designed for de-novo assembly because of the short read lengths.

Perhaps your question more revolves around assembly of very short reads?

Or perhaps your question revolves around the mis-assembly due to repeat regions?

In any case please restate your question more coherently.