Information Content of Sets of Biological Sequences Revisited

摘要: To analyze the information included in a pool of amino acid sequences, first approach is to align estimate probability each occur within columns aligned sequences and combine these values through an “entropy” function whose minimum corresponds absence information, that is, case where has same occur. Another alternative construct distance tree between (issued by alignment) based on sequence similarity properly interpret topology so model evolutionary property residue conservation. We introduce concept “evolutionary content” demonstrate at what extent more classical notion “information approximates new measure manner contributes sharper for detection protein binding sites.