Kmasker

FAQ

• Is there a tutorial or data set to get assistance on how to use Kmasker plants?

  We provide hands-on training for Kmasker plants available as DOI (https://doi.org/10.5447/ipk/2019/17). It includes input and output data as well as all commands used to generate these. Updated versions of the command files are available on our GitHub project page under the examples section. In addition, the GitHub repository includes documentation of the tool (https://github.com/tschmutzer/kmasker).

• Kmasker did not create figures?

  To avoid inconvenient run times, plotting is deactivated for datasets that contain >100 sequence IDs. Use a subset of your input data or use a local installation of the Kmasker plants command line version.

• I don’t receive results?

  This can have multiple reasons. Contact us for details. BUT there are some reasons that were observed a few times causing failures. First, please do not upload sequences in Microsoft Word documents (*.doc or *docx) into Kmasker. Sequences have to be in FASTA format. Second, only the four base characters ‘A’, ‘C’, ’G’ and ’T’ and additionally ‘N’ are permitted as characters in your input sequence. Make sure your input is fulfilling this criterion.

• I used Kmasker for primer screening and did not get any sequence as output?

  Please adjust the threshold for sequence length of the output. The default (100 bp) might discard your primer sequence.

• Is my data trimmed to the maximum input length (100 Mb)?

  No, please adjust your input before running Kmasker.

• Does the computed frequency needs to be normalized to the level of sequencing?

  No, the frequency values in the OCC files are already normalized to the corresponding WGS sequencing depth.

• What are the ‘X’ characters in the output file?

  All positions with frequency values larger than the ‘repeat threshold’ parameter will be masked by ‘X’ in the output file. In consequence, the setting of the repeat threshold (default: 5) is directly influencing the masking. Large repeat thresholds will decrease the number of X since only highly repetitive regions will be masked. On the other hand, if small values are used the proportion of X will increase.

• How should I use the parameter ‘length threshold’?

  This parameter is used as a filter after the index has been applied to your sequence and a frequency is computed for all sequence positions. The filter selects regions from the complete set of positions with a low frequency that are larger than the provided length. The shorter the ‘length threshold’ the more results will pass the filter. If you require long sequences (e.g. for FISH) you might increase this threshold.

• Is there a limit I can use when setting the size threshold?

  No, the limitation is the length of your input sequence.

• Do large ‘length thresholds’ increase the error rate?

  The error rate, in general, is not affected by the length threshold. You might just not get a sequence output if your setting is too ambitious.

• Which repeat libraries are used for annotation?

  You can provide FASTA sequences including your individual curated libraries of repeats. Kmasker will perform a blastn search and assign this information to repeats detected in your input data (see resulting GFF file). This option only is available in the local command line installation of Kmasker plants.

• How can I cite Kmasker plants?

  If you use most recent methods published in 2019, please cite Beier et al. 2019 (The Plant Journal):
  Link to publication: https://onlinelibrary.wiley.com/doi/10.1111/tpj.14645
  DOI: https://doi.org/10.1111/tpj.14645
  
  Original methods are described in Schmutzer et al. 2014:
  Link to publication: https://www.karger.com/Article/FullText/356460
  DOI: https://doi.org/10.1159/000356460