v1.3.2

Model update (#5)

• Change from minimap2 to winnowmap

• winnowmap performs better in repeat regions as shown in the paper https://doi.org/10.1093/bioinformatics/btaa435

• Add new error catch

• Add script to check if magnipore positions are within annotated regions

• add (0-based) information to readmes

• changed default parameters for winnowmap

• fixed bug that doubled the read counts

• minor bug fixes

• fix tests

• remove settings force_rebuild to true if segmentation is not done

• Update test according to new default mapping parameters

• add file paths to output

• add file paths to output
  main (#5)

v1.3.1

Changed

• the 7mer context for r9 to a 5mer context
  • some data showed significant signals near a mutation that are within a 7mer context, but not a 5mer context - hard to say if from mutation or not (no proof). Therefore, stick to the number of nucleotides measured together depending on the pore-type.
• MeanDiffAvgStdDev plots are now called MeDAS, meaning Mean Distance Average Standard-deviation
  • MeDAS plots showing the coverage
  • plots are now created using the magniplot.py
  • before: created with magnipore.py
• change part of logger file name created by the subscript nanocherlock.py from "magnipore" to "nanosherlock"
• gzip will now overwrite already existing output files
• summary output
  • typo fixes
  • less redundant text
  • more information about number of positions
• replaced minimap2 with winnowmap
• changed winnowmap parameters to the default values used by the ont_aligner toolchain from Oxford Nanopore Technologies

Added

• different ranges for the classification of a significant signal into mutation or potential modification depending on the pore type
  • r9: 5mer context
  • r10: 9mer context
• magnifilter.py: a script to filter the given .magnipore output file for a given coverage threshold
• magnicheck.py: a script to check if magnipore found the positions in a context of given validated ground truth positions
• multiprocessing of magnipore model comparisons
• Error for empty f5c file
  • 027: f5c eventalign file is empty

v1.3.0

Changed

• the 7mer context for r9 to a 5mer context
  • some data showed significant signals near a mutation that are within a 7mer context, but not a 5mer context - hard to say if from mutation or not (no proof). Therefore, stick to the number of nucleotides measured together depending on the pore-type.
• MeanDiffAvgStdDev plots are now called MeDAS, meaning Mean Distance Average Standard-deviation
  • MeDAS plots showing the coverage
  • plots are now created using the magniplot.py
  • before: created with magnipore.py
• change part of logger file name created by the subscript nanocherlock.py from "magnipore" to "nanosherlock"
• gzip will now overwrite already existing output files

Added

• different ranges for the classification of a significant signal into mutation or potential modification depending on the pore type
  • r9: 5mer context
  • r10: 9mer context
• magnifilter.py: a script to filter the given .magnipore output file for a given coverage threshold
• magnicheck.py: a script to check if magnipore found the positions in a context of given validated ground truth positions
• multiprocessing of magnipore model comparisons

v1.2.7

Added

• add missing xlabel to meanDiffStdAvgDist plot

Changed

• change xlabel from "Mean Difference" to "Mean Distance"
• reduce plotting data
• plot max 1000000 random positions to reduce runtime and prevent crashing (sometimes kernel killed the process)

v1.2.6

Changed

• fixed the following bug in function mapFast5Files of nanosherlock.py
  UnboundLocalError: local variable 'ridx' referenced before assignment

v1.2.5

Changed

• Fixed infinite range error in plotScores for TD Score and KL Divergence histplot

v1.2.4

Changed

• the sequencing summary is not mandatory anymore (must not be next to FASTQs)
• sequencing summary can still be provided to speed up indexing of fast5 files

Removed

• error code for missing sequencing summary

v1.2.3

Removed

• pod5 error, as it is installable via conda now

Added

• Error Code 15 the number of provided reference sequences it not equal 1 or 2
• 1 if both samples use the same reference
• 2 if each sample has its own reference
• more than 1 reference sequence per sample is still not supported

Changed

• READMEs
• does not use mafft anymore if both samples have the same reference
• mafft introduced gaps, if both samples have the same reference, the alignment should have 100% identity

v1.2.2

Fixed Bugs in Code

• when providing the same reference sequence errors occured in magnipore.py magnipore methd

updated conda description

v1.2.1

Remove broken pypi version