De Novo Assembly of Illumina reads with Velvet (1.0.17) and read alignment with Bowtie (0.12.7) | molecularevolution.org

I chanced upon this nicely written tutorial and also the idea to use bowtie to estimate insert size distribution! Neat! Also includes a little bit of R too

Expected learning outcome

The objective of this activity is to help you understand how to run velvet in general, how to accurately estimate the insert size of a paired-end library through the use of Bowtie, the primary parameters of velvet, and the process involved in producing a de novo assembly from Illumina reads.

Type the following commands in R to calculate mean and standard deviation, and draw a histogram of the insert size distributions:

sizes <- as.numeric(readLines("300bp_pe_def.sizes"))
mean(sizes, na.rm=TRUE)
median(sizes, na.rm=TRUE)
sd(sizes, na.rm=TRUE)
hist(sizes, nclass=50)

sizes <- as.numeric(readLines("3kb_mp_def.sizes"))
mean(sizes, na.rm=TRUE)
median(sizes, na.rm=TRUE)
sd(sizes, na.rm=TRUE)
hist(sizes, nclass=50)

When you are satisfied, you can quit R by typing q(). It is interesting to see now that the median differs significantly from what was estimated in the laboratory originally.
http://molecularevolution.org/resources/activities/velvetbowtie_activity