Click for Instructions
WebACT
previous 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | next

Generating and viewing a comparison from your own sequences

To access the form for data entry click the 'Generate' tab of the menu.

WebACT requires a minimum of two and a maximum of five sequences to generate comparisons. Once the number of sequences required is selected, these sequences can be entered in a number of ways:

  • pasting data in EMBL or FASTA format into the text boxes.
  • uploading a file in EMBL or FASTA format by using the 'BROWSE' field of the form.
  • entering an EMBL ID or accession number.
  • a combination of the above

When entering an EMBL ID or accession number, WebACT will retrieve the sequences and display the resulting comparisons fully annotated; similarly when pasting annotated sequence, the comparison will be displayed with the sequences fully annotated.

You are offered the option to receive an email notification when the job is completed; for large sequences we recommend entering a valid email address as comparisons can take some time.

a) Changing BLAST parameters

For generating both BlastN and TBlastX comparisons we have set BLAST options to the program defaults.

To change any of these simply click [show] on the form to view options and change as required (see below).


General Blast Options

Comparison Type: WebACT allows the use of two different blast programs for generating the comparison. BLASTN carries out a comparison between the sequences at a nucleotide level, whereas TBLASTX comparisons are carried out between six-frame translations of each sequence i.e. at the protein level. level. TBLASTX comparisons are considerably slower to generate than BLASTN comparisons, but are more sensitive, while BLASTN comparisons will better indicate highly conserved regions between similar sequences. We would recommend initially carrying out a BLASTN comparisons, and if this does not give satisfactory results, then carrying out a TBLASTX comparison.

Filter Sequence: The presence of regions of low complexity or sequence repeats can result in hits with high scores which do not reflect sequence similarity. Filtering the sequences will mask out regions of low complexity in the query sequence prior to carrying out the comparison. If a low number of regions of similarity are found by your comparison, it may be worthwhile repeating the comparison with filtering turned off.

E Value: A BLAST E-value is used to report the significance of each hit, and represents the number of hits which would be expected to occur by chance in a database of given size. Altering the E-value cut-off for the comparison allows the stringency of the comparison to be adjusted. Decreasing the e-value will allow less hits which are likely to have occurred by chance into the results, while increasing it will have the opposite effect.

Cost To Open A Gap: When calculating the score of an alignment, BLAST penalises the creation of a gap. Reducing this value will favour the introduction of gaps into the alignment, which may be of benefit for more distantly related species.

For BLASTN comparisons, the default penalty is 5, whereas for TBLASTX, the default changes according to the matrix employed. See the table below for default values.

Cost To Extend A Gap: When calculating the score of an alignment, blast penalises each extension of a gap which is required. This parameter is typically smaller than the cost to initially open the gap, favouring the creation of larger gapped regions over multiple short gaps.

The default value is 2 for BLASTN comparisons, and varies according to the selected matrix for TBLASTX comparisons. See the table below for further information

Default Penalties for Different Matrices
Matrix Gap Creation Penalty Gap Extenstion Penalty
BLOSUM45 14 2
BLOSUM62 11 1
BLOSUM80 10 1
PAM30 9 1
PAM70 10 1

BLASTN Options

These options will only be displayed when the 'Comparison Type' in 'General Options' is set to 'BLASTN'

Word Size: Blast initially splits your query sequence into 'words' of a defined length, and searched the other sequence for exact matches to these words. This parameter defines how long these words are (in bases). The default word size for nucleotide comparisons is 11. Reducing this will produce more hits in the results, but with an increase in the computation time required. Increasing the word size will increase the length of the average hit, and make the search faster to run, but produce less hits overall.

Penalty for Nucleotide Mismatch: This scoring penalty is applied to every mismatched base in an alignment, consequently the greater the number of mismatching bases, the lower the score of the alignment. The default value is based on sequences with 99% similarity.

Reward for Nucleotide Match: This score is appled to every matching base in an alignment. As for the Nucleotide Mismatch Penalty, the default value is based on sequences with 99% similarity.

TBLASTX Options

Word Size: Blast initially splits your query sequence into 'words' of a defined length, and searched the other sequence for exact matches to these words. This parameter defines how long these words are (in bases). The default word size for TBLASTX comparisons is is 3. Reducing this will produce more hits in the results, but with an increase in the computation time required. Increasing the word size will increase the length of the average hit, and make the search faster to run, but produce less hits overall.

Codon Translation Table: Defines which codon translation table to use when carrying out translations prior to the comparison. Select the most appropriate value for the source of your sequences. If there is no obvious code appropriate to your sequences, leave at the default setting, which will be fine for the majority of cases.

Scoring Matrix: The scoring matrix assigns a score for each pair of aligned amino acids. A number of different matrices are available, although in the majority of cases, the default BLOSUM62 matrix is the best choice.

b) Submitting Your Search

When sequences / IDs have been entered, click 'Submit Query' and WebACT will process your comparison.

You are given a URL that can be bookmarked and returned to once the job is complete or you can wait untilthe results are returned to the browser. Should your sequences contain no BLAST hits above the defined cut-off level you will be informed.