Patchy alignment in bowtie

Question

[second_exon]

I am biologist, learning bioinformatics, please bear with me, if I go wrong in basics.

I assembled mi-seq reads ranging between 33-340bp using velvet. I tried different K-mer values ranging between 31-43. But, bowtie alignment reports, only 0.01% of reads were aligned, but the final contig length was around 14 to 40kb across different K-mer. I also tried assembling short reads (-short) in velvet but nothing has improved.

I am very confused and where I am going wrong?

Thanks in advance