New Features and Changes
- The software detects the orientation of index sequences during demultiplexing instead of using the orientation specified in RunManifest.csv. If the detected orientation differs from the run manifest, the statistics output file indicates the detection. This feature creates a more robust method of demultiplexing by reducing user errors.
- A new statistics file, IndexingAssignment.csv, reports the indexing metrics for each sample pair.
- A new optional argument, qc-only, reports statistics without generating any FASTQ files.
- The software now supports rclone, a command-line program that manages files in cloud storage, as input and output remotes to interact with cloud storage from various providers. Element has not tested all rclone configurations.
Bug Fixes
The following bugs are fixed:
- When executing Bases2Fastq on a subset of data, the complete dataset was downloaded, and then slices were operated on. To optimize download time, only the subset data that option filters provide is downloaded, reducing download time.
- If RunParamters.json is missing FlowCellID, Bases2Fastq reverts to using the RunID as the flow cell identifier to prevent blanks in FASTQ headers and unidentified flow cells.
- If RunParameters.json is missing RunName, Bases2Fastq reverts to using the RunID as the run name to prevent blanks in FASTQ headers.
- If whitespace was added to any fields in the FASTQ header, the whitespace is trimmed to prevent processing errors in the FASTQ headers.
Known Issues
When one one of either read is greater than 48 cycles and the other read is less than 32 cycles and Paired-End type adapter trimming is used, bases2fastq will fail during execution. As a workaround, Single-End type adapter trimming can be used instead.