I've run Blast normally with default parameters. Alignment scores are give in fractions like 84.3. But the BLOSUM 62 matrix contains no fractional scores for any amino acids. So I would like to know how those fractional score is arrived. http://www.ncbi.nlm.nih.gov/protein/9501247?report=fasta
•
written
by
barathvajramachandran •
1 answer
http://homepages.ulb.ac.be/~dgonze/TEACHING/stat_scores.pdf
http://www.ncbi.nlm.nih.gov/blast/tutorial/Altschul-1.html
You have to remember that BLAST is not a simple sequence aligner, it tries to find you the likely best hit(s) as quickly as possible. This means BLAST trades accuracy (not precision) for speed, which means it uses many heuristics to speed the search of the sequence space up.
The score isn't an actual alignment score, but a transformation on that score to normalize on the size of the search space and the scoring system used.
If you just want a raw score, you shouldn't be using BLAST.
Log in to answer this question.
-
What are recommended parameters for the local protein alignment ?
written by Alexander •
-
Setting up local blast in webserver
written by sharmatina189059 •
-
B Cell Epitope Prediction: Maximizing Antigenicty and Minimizing Sequence Length
written by jondriscoll1988 •
-
How to explain difference between two different calls to PSI-BLAST?
written by a.nash •
-
blastp word_size parameter seems to be ignored - BLASTP 2.4.0+
written by matthys.potgieter •
-
How Is The Word Score For The Look-Up Table Determined For Blast?
written by Niek De Klein
-
Substitution Rate Calculation From Protein Alignment
written by Raghul
-
All Pairs Similarity Search Under A Substitution Matrix For A Large Number Of Short Protein Sequenc…
written by Harish Tella •
-
How Log-Odds Scores For Amino Acid Substitutions Matrices Should Be Calculated?
written by Jan Kosinski
-
Score Protein Variants Based On Frequency Of Aa In Multiple Sequence Alignment
written by Tim
