Cloning bias is a tendency of certain regions of the genome to be cloned less often than others during sequencing, and thus less likely to be sequenced. This results in a lower expected read coverage in these regions, which magnifies the gap-producing effects of random chance. To get enough data to sequence these regions, a higher overall coverage is needed. The most prominent cause of cloning bias is AT-richness, which affects cloning rates in the Sanger sequencing process. In 454 sequencing process, another kind of cloning bias causes homopolymers to be counted incorrectly.

Thanks alot Daniel Clark! Your info was very helpful!!

As for the _sources_ of cloning biases, some sequences, by chance, contain signals that interfere with the biology of a particular cloning vector (they may contain an inverted palindromic sequence which causes the replication of the insert in that cloning vector to stall, among other causes); some cloning vectors have specific preferences for sequences at the ends of the insert, which may be more common in some regions of the genome than others; some cloning vectors have preferences for sequences in a given range of lengths, which may be more difficult to produce from some regions of the genome than others.

Thanks Max!