How to interpret BLAST E-values and scores?

Understanding BLAST E-values

• The E-value, or Expect value, is a statistical measure used in BLAST to estimate the number of hits one can expect to find by chance when searching a database. A smaller E-value indicates a more significant match.

 

• The E-value depends on the size of the database, where larger databases can increase the E-value for the same alignment score. It reflects the background noise of a database search.

 

• An E-value of 1 implies that you might expect to find one match with this score by chance. The lower the E-value, the less likely it is that the match is due to random chance.

 

Understanding BLAST Scores

• The score in a BLAST result represents the quality of the alignment or match between the query and the database sequence. It is calculated using a scoring matrix that includes positive and negative values based on match/mismatch and gaps.

 

• A higher score indicates a more reliable and accurate alignment. Scores directly affect the calculation of the E-value; the higher the score, the lower (better) the E-value.

 

• Score units are often measured in bits. Therefore, scores can be compared cross searches or databases if they use the same scoring system.

 

Interpreting BLAST Results

• Begin by examining the E-value in the result. Focus on alignments with the lowest E-values, as they suggest the most significant matches.

 

• Consider both the E-value and the alignment score when assessing the significance of an alignment. A high alignment score with a low E-value is generally the most biologically relevant match.

 

• Be cautious of alignments with very high E-values. These could be spurious matches that occur by chance and may not indicate a true biological relationship.

 

• Acknowledge that database size affects E-values: as databases grow, expect E-values to rise for the same score. Always interpret results in the context of the database's size and content.