- Version 2.10 (Aug 18, 2020
- Check the upper-bound of quality score values and throw an error when the maximum allowed quality score is exceeded.
- Reordered the change log to display the newest changes first.
- Added a try/catch block in the trim() function to print thread-level errors.
- Version 2.09 (Aug 17, 2018)
- Added informative error messages to fasta.cpp and FaQCs.cpp to help users diagnose corrupted input files.
- Version 2.08 (May 1, 2018)
- Fixed a bug due to the "--polyA" command line argument expecting an argument (it should be a boolean flag).
- Version 2.07b (Mar 16, 2018)
- Fixed a bug in the "vector& operator+=()" function in trim.cpp that improperly truncated the length histogram (as displayed in the output plots). It was possible to completely erase the length histogram, which lead to an R script error (when the R function "read.table()" was passed an empty intput file).
- Fixed a bug in the "matrix& operator+=" function in matirx.h that improperly tuncated a matrix, leading to a missing file and R script error (similar to above bug with the vector += function).
- Version 2.06 (Jan 9, 2018)
- Fixed bug that caused an error to be thrown when a negative quality score is encountered. As mentioned on the FASTQ wikipedia page (https://en.wikipedia.org/wiki/FASTQ_format), the old solexa+64 format does allow for negative quality scores. As a quick fix, all quality scores are now clamped to be greater than or equal to zero. The storage formats for the quality offset values (Options::input_quality_offset, Options::output_quality_offset and Options::quality) have been changed to signed char values (previously these were unsigned int values).
- Version 2.05 (Oct 18, 2017)
- Fixed bug in R script that generated incorrectly normalized nucleotide content plots (thanks Chien-Chi!).
- Version 2.04 (Oct 17, 2017)
- Fixed bug that generated incorrect plots when 5' and/or 3' trimming (i.e. --5end or --3end) was used. Also corrected bug in the output R script implementation that passed the same quality matrix to the before and after graphs.
- Version 2.03 (Oct 2, 2017)
- Fixed bug that caused an error to be thrown when the input quality scores were in Illumina 1.5+ format (score offset == 64).
- Version 2.02 (July 12, 2017)
- Fixed bug that did not property parse fastq deflines for paired-end data. This caused a spurious error message about read 1 and read 2 not matching. Following the FaQCs.pl script, terminial ".\d$" and "/\d$" are now removed prior for testing if a pair of reads has a consistent header.
- Version 2.01 (May 12, 2017)
- Fixed the bug that prevented '-t' from being recognized. The '-t ' now functions the same way as the Perl version of FaQCs.pl