Web Input

Why is email needed?

A valid email address is needed in case you request that the results be sent.

How to name a project?

Your project title is required to name this bipad run. The title will also be used as the title of the sequence logo generated on the fly if your result display is online suppose that you are a registered user.

What is a two-block pattern (bipartite) search?

A search pattern can be either one-block or two-block (bipartite). You can set one pattern each run. Note that the search pattern specified is your hypothetic model to be built by Bipad. It should be based on your biological observation or published information related to the problem you are looking at. We recommend that you try a little wider motif than your thought and then progressively refine your model by adjusting your motif widths and gap ranges based on the sequence logo or bipad output you generated. The Bipad web server currently does not automate this refinement procedure. Therefore it is up to the user who controls the searching scheme. Obviously an improper search pattern may lead to a biologically meaningless motif model.

A two-block (bipartite) pattern (see illustration below) is an independent functional unit on the upstream or downstream of a co-regulated gene, which is often recognized by a homodimer or heterodimer protein complex. A two-block search pattern (bipartite) can be expressed as JL<[dmin, dmax]>JR.

Widths: JL, JR, the widths of the first and second motifs respectively. The minimum width allowed is 4 bps and the maximum width is 23 bps.
Gaps: dmin - minimum gap, dmax - maximum gap size. The minimum gap allowed is 0 and max is 20 bps.
The limitation could be relaxed based upon request.

A one-block can be thought of as zero-gap two-block pattern where two motifs are merged into one.

If it is a single-block, then the JL is picked as a single-block motif width and JR, dmin and dmax are ignored. If it is a two-block, you are required to set all values of JL, JR, dmin and dmax.

BiPad spends about double time on a two-block pattern (bipartite) search than on one-block motif search.

What is an alignment mode?

Bipad outputs motif models based upon optimal multiple local alignment which may be performed either on forward strand or both strands. The alignment mode should be in agreement with your biological problem or hypothesis: motifs may be located on one strand only or either strands. It takes bipad more time with both strands aligment to complete your job than for single strand alignment mode.

Number of runs

The default is 10 runs, however, you can specify more runs if needed. The maximum number of runs allowed is 1000 due to limited web server capacity. If more than maximum is needed, please contact us: cbi at cmh.edu. Remember that the more runs you set, the longer you need to wait for your results.

Fine tuning of width

The default is tune-off. You'd better keep it off when starting a new problem in order to display the logos in graphic. If it is on, then no logos will be showed in the output. If switch on fine-tuning, the specified width will become the core, Bipad will make a batch of runs and draw a UII graph. The maximum tuning width is set to 24 bps.

DNA sequences: FASTA format

To enter your own sequence, you can either copy and paste into the text area or type in directly. Sequences should be consisted of {A,C,G,T, 1-9}. You can use number to count the rows of nucleotides. Upper and lower case are equivalent, and characters other than {A, C, G, T and 1-9} are illegal and may cause bipad program halted. Sequences can also be uploaded as FASTA file format.

Here is the FASTA format example:



The maximum length of a sequence is 5000 bps and the maximum number of sequeces each time allowed is 2000. The maximum total length of the sequence set is 10,000,000 bps. Your sequences are kept in our server for one day. These limitation could be relaxed upon your request. Notice that if you run bipad standalone program, all the limitations are gone. However, in command line bipad program only accepts sequences in text format, that is, sequences are separated by returns .

Sequence model: OOPS or ZOOPS?

The default sequence model is OOPS (one two-block site occurence per sequence in the dataset). You also could choose ZOOPS (zero or one two-block site occurence per sequence) as an option.

How many bases flanking a motif?

Residues upstream and downstream of motifs. Flanking bases are in lowercase and motifs are printed in uppercase. Flanking bases will not be showed in a sequence logo due to map size limit. If your flanking bases are out of bound, bipad takes the boundary values which results in printing a full alignment. As defaults, there are no flanking bases.

What is output/result options?

Results can be displayed either online or sent to you by email depending on your options. The default is to display your result.