Net sequence gazing

Last month, The spider's web looked at the range of biological resources available on the net. This month there are more specific tips on how to get the most out of these tools, for the least effort, with an emphasis on tools for sequence analysis. As is true of almost any use of computers, getting something done on the net requires a lot of non-intuitive knowledge, and net knowledge becomes redundant at a fantastic rate. So, for the moment the best advice that can be given is to use a net search to find the internet tools you need. Surfing, whether literal or metaphorical, is a great leisure activity but not an especially productive one. It is nearly always most efficient to use internet search facilities to track down what you need. New internet search engines are appearing all the time, each claiming to be faster and more comprehensive than the last. Alta Vista and OpenText have recently joined the stalwarts, Infoseek, Lycos and Yahoo. If you want to find a tool to determine which conserved domain motifs are present in your protein sequence, try a search for 'protein conserved domain'. This will find, amongst other things, a great new tool, called Beauty, developed for exactly this purpose. Whichever search service you rely on, it pays to become familiar with its more advanced search facilities. For example, at Alta Vista, putting a + sign in front of a word makes it essential, whereas putting a – sign causes documents matching that word to be omitted. At Infoseek, to make it easier to search for a person by name, the engine conveniently guesses that two adjacent capitalized words should be searched for as a phrase. Molecular sequence analysis For sequence-gazers, things have been immeasurably improved by the development of systems that combine many different tools and databases into a common framework. Entrez, developed at the National Council for Biotechnology Information, works from a non-redundant database compiled from all the main international protein and DNA sequence databanks, together with a fair subset of Medline bibliographic records. The indexing process uses the BLAST algorithm to measure similarity between all the sequences in the database, allowing users rapidly to retrieve the 'nearest neighbours' of any given sequence. Novel sequences can also be compared with existing databases using BLAST. Other advantages of Entrez include extensive cross-linking and plenty of control over the format of reported …