Introduction to sequence alignment  (Page 2/4)

Under program selection, the user can choose to optimize blastn for the type of sequence search that best relates to their problem, for highly similar sequences (megablast), more dissimilar sequences (discontiguous megablast)or somewhat similar sequences (blastn). At the bottom of the web page, there is a link that allows the user to view the default values for formatting (return the descriptions for the top 100 scoring alignments) and for the algorithm parameters. If you selected megablast, optimize for highly similar sequences, the word size used in the search will automatically default to 28 nucleotides. If you select blastn to optimize for somewhat similar sequences the word size will be 11 nucleotides, the historically common default value for blastn.For the purposes of this example, select to optimize for somewhat similar sequences (blastn) and click on the BLAST button to submit your query to the BLAST queue.BLAST should state that the query was successfully submitted, and provide a request ID that accesses the results. The page updates itselfautomatically, you do not have to refresh it.

The results begin with a graphical overview that uses the query sequence as therule, shown separately at the top of the graph. In this example, the rule (our query sequence) is 453 nucleotides in length. The score of each alignmentis divided into one of 5 ranges by the five different colors in the color key provided, and the alignments are placed in the order of best scores first. Thescores are normalized, and normalized scores are given in units of bits, to allow comparisons between alignments. More information on scores can andshould be perused in the Altschul tutorial on the NCBI website. Clicking on any of thefirst 50 lines in the graph takes the user to the associated alignment, located in the last section of this resultspage.

The second section of the results page, immediately following the graph, contains the top 100 alignment scores and their sequence descriptions. Here, the "hits", or matched sequences returned as a list of results, contain links to their GenBank entries, their GenBank definitions under the column entitled "Accession", the normalized scores ofthe alignments, and the E-values. The E-values quantify the probability that the listed alignment might occur randomly. The scores are linked to the actual alignment further down the page. Clicking on a score will take you directly to the alignment that obtained that score.In some cases, there are icons containing capital letters in the links column, which are links to related entries in one of the NCBI databases, such as UniGene unique gene cluster information , Entrez Gene and GEO gene expression and hybridization array data . Look at the first 5 "hits" in the results list.