All videos › Biology › Grade 12
Genomics - Comparing Gene Sequences With BLAST You have read a gene from a new bacterium. Is it like any gene already known? See how a sequence database search compares it, and how to read percent identity, query cover and the E-value.
Sign in to save your place
Biology · Grade 12 · Microbiology and molecular biology · 3.5 minutes · UK English
Transcript 0:02 Imagine you have read a gene from a bacterium nobody has0:06 studied. Is it like any gene that is already known?0:10 You already know that DNA is a sequence of four bases,0:14 A, C, G and T, and that a gene is a0:16 stretch of that sequence. In this video,0:19 you will use a sequence database search to compare one gene0:24 with many others, and learn to read the result.0:28 Bioinformatics uses computers to store, analyse and share biological data,0:33 such as DNA sequences. One public database is GenBank,0:37 held by the US National Institutes of Health.0:41 A tool called BLAST takes your sequence,0:44 called the query, and compares it with the sequences in a0:48 database. It finds similar regions and works out how significant each0:54 match is. A close match can suggest that two genes share0:58 a job, or a common ancestor.1:01 Here is a worked example with a made-up gene.1:05 BLAST lines up twenty bases of your query against the top1:11 hit. Query, A T G G C C T T A1:14 G C A G G T C T G A A.1:16 Hit, A T G G C G T T A G1:18 C A G G T C A G A A.1:20 Position six has a C in the query and a G1:24 in the hit. Position seventeen has a T against an A.1:29 Every other pair matches. So twenty minus two gives eighteen matching1:36 positions. Percent identity is matches divided by aligned positions,1:42 times one hundred. Eighteen over twenty is ninety percent.1:48 Each hit also has an E-value. It is the number of1:52 matches this good, or better, that you would expect by chance1:56 alone in a database of that size.1:58 So a lower E-value means a more significant match.2:02 The E-value also depends on the size of the database.2:06 The hit list also gives query cover,2:08 the percent of your query that is included in the alignment.2:13 Your turn. Your search gives two hits.2:16 Hit A has an E-value of 5.2:18 Hit B has an E-value of 0.001.2:21 What does the smaller E-value of hit B tell you?2:25 Is it, B is weaker, because its value is smaller?2:29 B must be from the same species?2:32 Or, B is less likely to be a chance match?2:36 Pause the video and decide.2:47 The answer is the third choice. B is less likely to2:51 be a chance match. Five matches as good as hit A2:54 are expected by chance, which is five thousand times more than2:59 for hit B. A low E-value does not tell you the3:02 species, and smaller does not mean weaker.3:05 Now a common mistake. Do not trust percent identity alone.3:09 A hit can be one hundred percent identical over only a3:13 short part of your query. Always check query cover as well.3:18 Here is one tip to keep. Read identity,3:21 cover and E-value together, never one alone.3:25 A strong hit has high identity, high cover,3:29 and an E-value close to zero.3:32 Let us recap. Bioinformatics uses computers and databases to store and3:37 compare sequences. BLAST compares your query with a database and lists3:42 the hits. Percent identity is matching positions divided by aligned positions,3:48 times one hundred. A lower E-value means a match is less3:52 likely to be chance. Query cover shows how much of your3:57 query lines up. Thank you for learning with Worked Example School.Spotted a mistake? Tell us at contact@workedexampleschool.com .