Comparative analysis is one of the foremost tools in biology. The most common comparative method is sequence alignment, which can be performed in many ways. By using any sort of alignments, competing sequences are evaluated by scoring schemes for estimating similarity. The effectiveness of database searches is dependent on a large number of correlative factors. There are many equivalent DNA and protein sequence databases. It may be useful to restrict the search to a main database.

In searching for similarity, the amino acid sequence derived from the DNA sequence is more informative than the DNA sequence. When proteins are analysed, they are divided into their compact globular structures, such as domains. Domains are used as the units of protein classification. Proteins are also clustered into families and superfamilies whose members have diverged from common ancestral forms. There are other types of protein classification that consider both sequence similarity and structural similarity. Based on these similarities, multiple databases have been constructed with different focuses.

Most protein families can be characterised by sets of locally similar sequence segments, referred to as patterns and motifs. The detection of homology between a newly-determined sequence and a sequence motif in a protein databank is often the most important clue to the function of a gene

This thesis summarises the identification and characterisation of a novel mammalian gene family. As a common strategy for the genes presented here, full-length cDNAs were isolated, sequenced, the genomic structures characterised and the expression patterns analysed. Moreover, the existence of homologous genes in databanks extends the family to other eukaryotic organisms. The detection of homology between the gene families suggests a functional equivalency among the gene families. By complementation studies in yeast mutants, we found that homologous genes from a mouse gene family could rescue the yeast mutants lacking the ability to synthesis very long chain fatty acids and sphingolipids.

In conclusion, this study presents the identification and characterisation of a gene family in which mutations can have pathogenic consequences and provides evidence that the members of this mouse gene family participate in the process of very long chain fatty acid synthesis.