BEAM discovers novel, ungapped motifs (recurring, fixed-length patterns) in RNA secondary structures starting from RNA sequences or structures. BEAM Web Server can fold RNAs with RNAfold (MFE) or RNAstructure (MEA). Alternatively, the server accepts user provided structures as well as BED files with genomic intervals. See this guide for instructions on input format.

you may experience issues with our web server in this month because we are undergoing massive server relocation

- Now Drosophila (dm6) is available for the BED to BEAM pipeline. Check it out (use only with CDS and no filtering options)
- It is possible to use the whole BED file (without mapping to any region) by using the no filtering option)

Pre-computed examples:
Sample output from these sequences
Another example from these SLBP-targets (GSE62154)
Another example from these LIN28A-targets (DoRiNA - see main paper in the docs)
An example from a BED file: U6/atac snRNA from Rfam

Upload RNAs

The minimum number of submitted sequences is 4.
The maximum number is 500.000 (see table for folding limitation).
Software Requested Max #RNAs Max RNA length
RNAfold 100000 500
500 2000
RNAstructure-MEA 50000 500
None (input already has dot-bracket) 500000 2000
input sequences are accepted in multiFASTA format:
  • The line containing the name and/or the description of the sequence starts with a ">";
  • The words following the ">" are interpreted as the RNA id;
  • The following line should be the RNA nucleotide sequence;
  • The subsequent lines are interpreted as secondary structure information (Optional)
  • Please consider that any duplicate ID in the fasta file could cause unexpected errors.
The IUPAC notation is accepted for nucleotides (case-insensitive).
Please consider a run-time roughly equal to 1.5 minutes per motif in standard conditions (*** M*R*S*10-4 minutes, where M is the number of motifs requested, R is the number of random starts, S is the number of Total Steps, by default 15000). To get the full wait-time, if structures are not provided, add the time needed by the chosen secondary structure predictor (for RNAfold usually max 1 minute per 1000 sequences. RNAstructure takes longer), see the limits table and graphs for detailed information).


File:
(Optional) Load a background:


Upload BED

The minimum number of submitted sequences is 4.
The maximum number is dependent on the options requested (as per limits table in the documentation).
input sequences are accepted in BED format:
  • Be aware that the effective length of the uploaded RNAs will be
    interval size + 2*EXTENSION
  • Please remove any comment line and trailing blank lines before submitting
Please consider a run-time roughly equal to 1.5 minutes per motif in standard conditions (*** M*R*S*10-4 minutes, where M is the number of motifs requested, R is the number of random starts, S is the number of Total Steps, by default 15000). To get the full wait-time, add the time needed by the sequence retrieval (bedtools pipeline in the docs) and the secondary structure predictor (RNAfold usually max 1 minute per 1000 sequences. RNAstructure takes longer, see the limits table and graphs for detailed information).


File Bed:


 EXTENSION:
REGION:
GENOME:
(Optional) Load a background:



BG REGION:
BG GENOME:

Load an example


Or paste sequences




Options:

Min motif width:
(5-150)		Max motif width:
(5-150)		# motifs:
(1-10)		Fold:
+ Advanced

Advanced options are described in the docs

Annealing steps (1000-50000):
Total steps (1000-50000):
Starting temperature (50-300):
Model Limit (10-200):
Cooling rate (10-4-1):
Random starts (1-10):
Consider unpaired branch characters:
Keep all the runs (for Random starts>1):
Clean:
Add optional commands: