Overview of Help-Pages

RegionMiner: BED File Toolbox

Introduction

• converting from BED format to a different format:
  • from BED to genomic sequences
  • from BED to a UCSC Genome Browser custom track
• converting from other formats to BED format:
  • from GFF to BED format
  • from wiggle to BED format
  • from Illumina export.txt format to BED format
• pileup removal from a BED file
• extracting subsets from a BED file by criteria like length, chromosomal location, name, score, or strand
• extending the regions within a BED file by a certain number of basepairs up- and/or downstream
• sorting a BED file (e.g. for NGSAnalyzer input)
• comparing two BED files to find unique or overlapping regions
• concatenation of BED files
• mapping of sequence files to the genome to get BED files

General Parameters for the BED file toolbox

Available Actions for BED files

The parameter selection for each of the available tasks is hidden by default, click on the next to the section header of the desired action to see the parameters.
Note that the pivotal factor for the selection of the action is the radio button in front of the action within the parameter selection. "Upload to current project" is set as default action, if nothing else is selected.

Conversion to sequence

Tick the radio button to convert a BED file into the corresponding DNA sequences from the selected organism. To additionally extract basepairs up- and/or downstream of the regions, enter the number of basepairs to be extracted in the form.
The resulting sequence file is in FASTA format, with additional annotation as described in the Genomatix annotation syntax section. The file can be downloaded to the local computer or saved in the Sequences section of the Genomatix project management and can be used for other tasks that require sequence input.
Note: The extracted sequences will of course depend on the selected organism AND the selected ElDorado version (i.e. the genome build).

Conversion to UCSC Genome Browser custom track

The input BED file will be converted to a BED file with information for display in the UCSC Genome Browser (i.e. containing "browser" and "track" lanes). The resulting file can be saved locally and can then be added to the UCSC Genome Browser as custom track. It will be displayed as "User Track generated by Genomatix" in dark blue, starting the display at about the first region found in the BED file.

Note that Genomatix uses the designation 'chrMT' for mitochondrial chromsomes, while UCSC uses 'chrM'.

Conversion from GFF to BED format

To convert a file in GFF (General Feature Format) to BED format, a GFF file can be uploaded here. An additionally selected BED file in the Input section will be ignored.
The correct species for the input file must be supplied to allow checking correct chromosome numbering.

Conversion from wiggle to BED format

To convert a wiggle file (as described at UCSC) to BED format, a wiggle file can be uploaded here. An additionally selected BED file in the Input section will be ignored.
Note, that only "variableStep" and "fixedStep" wiggle files can be converted. Also, the uploaded file must contain the keywords "type wiggle_0" in the track line, otherwise they are not recognized as wiggle files.
The correct species for the input file must be supplied to allow checking correct chromosome numbering.

Conversion from Illumina export.txt to BED format

Pileup Removal

Subset of Regions

Select one of these actions to extract a subset of the regions within a BED file. The result will also be a BED file which contains only the regions from the input file that fulfill the set criteria.

• chromosomal position given by chromosome and positions on the chromosome
• the number of the region within the BED file, e.g. the first 100 regions
• the strand of the regions
• the region length (i.e. between x and y bp)
• the region name or ID (column 4 in the BED file) matches one of the given strings (case-insensitive!)
  With the option exact match the search string "ID_11" will find only "ID_11", where as substring match will also find "ID_112" and "thisID_11"
  Note that not all BED files contain names/IDs for regions since this is an optional column.
• the region score (column 5 in the BED file) between a lower (>=) and upper (<=) value
  The values can be given in scientific notation (e.g. 1e-16), which can be used to filter files from ChIP Clustering task
  Note that not all BED files contain scores for regions since this is an optional column, in this case score=0 is assumed.
• the regions with the highest/lowest scores

For the length and score criteria, the lower or upper value can also be set to arbitrary by entering a "-" into the corresponding parameter field. This allows e.g. to select all regions with a length >= 100 bp from the input BED file without knowing the maximum length of the regions.

The resulting BED file can be downloaded and e.g. used for other RegionMiner tasks.

Extension/Reduction of regions

Select this action to extend all regions of a BED file by a certain number of basepairs/positions in either one direction (up- or downstream) or in both directions. The resulting BED file contains all valid regions, which are extended by the given extent, unless the chromosome ends are reached (i.e. the positions are not smaller than 0, and not larger than the corresponding chromosome). The regions can be shortened by entering negative extensions.
Example: if a region on chr1, positions 100 to 200 is extended by 50 basepairs in both directions, then the resulting region is chr1, positions 50 to 250.

Sort regions

Here, the input BED file is sorted (chromosome alpha-numerically, positions numerically). A sorted input BED file can be of advantage for e.g. the NGSAnalyzer task in RegionMiner, as the performance will be enhanced especially for repeated runs with the same input.
The resulting BED file can be downloaded and e.g. used for other RegionMiner tasks.

Comparison of BED files

Here, two BED files can be compared to find unique or overlapping regions. Since we are comparing regions with a certain extension, parameters are necessary to define the "amount of overlap" that is used for analysis.
There are two parameters:

• Minimum overlap between regions
  This variable determines the minimum number of nucleotide positions that two regions must share to be counted as overlapping
• Maximum distance between regions (start-start/end-end)
  This variable determines the maximum number of nucleotides that may lay between the starts or the ends of two regions to be counted as overlapping

If one of the criteria above is fulfilled, two regions are counted as overlapping.
Examples:

1. region X from position 1000 to 2000, and region Y from 1900 to 3000 are overlapping by criterion 1, if the overlap variable is set to 50
2. region X from position 1000 to 1055, and region Y from 1050 to 3000 are overlapping by criterion 2 if the distance variable is set to 100
3. region X from position 1000 to 2000, and region Y from 2040 to 2090 are overlapping by criterion 2 if the distance variable is set to 100

The resulting table shows the number of

• Unique regions from the first BED file
  all regions that were only found in the first BED file, i.e. do not overlap with any regions from the second BED file
• Unique regions from the second BED file
  all regions that were only found in the second BED file, i.e. do not overlap with any regions from the first BED file
• Overlapping regions from the first BED file
  all regions from the first BED file that do overlap with at least one region in the second file
• Overlapping regions from the second BED file
  all regions from the second BED file that do overlap with at least one region in the first file
• Merged overlapping regions from both BED files
  all regions from the first and second file, that do overlap merged to one region

Each of the above subsets can be selected, and can be downloaded as a BED file or can directly be saved into the Genomatix project management for further analysis with other tasks.

Concatenation of BED files

Here, two or more BED files can be concatenated. Note, that you need to select the first file from the selection on top of the page, and the file(s) to add to the first files here. The files are simply concatenated while removing certain comment lines, but they are not sorted.

Mapping of sequence files to the genome to get BED files

If the mapping action is selected, a sequence file in FASTA or GenBank format must be supplied. The sequences are searched in the selected genome and - if found - the corresponding positions on chromosomes are written into a BED file. For finding the positions for mapping/alignent within the genome the same algorithm as for the sequence input in ElDorado is used.

The score of the resulting region in the BED file represents percentage of the input region that was aligned to the genomic sequence. Additionally, the alignment quality of the mapping is given in column 7 of the BED file (which is not used by other Genomatix programs).

There are two parameter settings:

• Extract max. one region for a sequence:
  if this option is checked, only the best of several mappings is used for the BED file output, else several mappings/regions may appear in the BED file for each input sequence.
• Sequence on genome may span at most x % of the input sequence:
  If this option is checked, the target region of the mapping is additionally checked for length, relative to the input sequence. Eample: to avoid long inserts, the percentage can be set to 150%, so that the genomic sequence is at most 1.5 times the length of the input sequence. But if e.g. mRNAs are to be mapped back, this parameter should be set to a large value (or the option should be deselected), to allow for large introns in the genomic sequence.

This task is limited to a certain number of input sequences.