would not be surprising for mice and humans to contain more like 50,000-100,000 such enhancers.

Other methods have been used to identify enhancers, based on the clustering of binding sites for sequence-specific transcriptional activators and repressors {see Chapter 18, Box 18-6). The recognition of regulatory sequences in DMA poses a much greater challenge than the identification of protein-coding sequences as regulatory sequences are not subject to constraints as stringent as that of the genetic code. Hence, it is likely that a combination of bioinformatics methods will be required to identify regulatory DNAs in whole-genome sequences.

The most commonly used genome tool is BLAST (basic local alignment search tool). There are variations in BLAST programs, but they all share the common feature of finding regions of similarity between different protein coding genes (Figure 20-21). There are many ways in which a BLAST search can be done. One involves searching a genome, or many genomes, for all of the predicted protein sequences that are related to a so-called query sequence. Consider the following example. We have already discussed the even-skipped (eve) gene in Chapter 18. The eve gene encodes a homendnmain protein that is essential for the segmentation of the Orosophila embryo. The Eve protein is composed of 376 amino acid residues. The homeodomain

Sx «■£>!• t sJ. 'I ï win 1(55 Nà*. tfl id H


MHI1. kuu

Was this article helpful?

0 0

Post a comment