Filter

Description

After running the annotate command on MAS-seq data, we expect that MAS-seq adapters will be found in sequential order throughout the length of the read. Reads that violate this expectation are potentially mis-segmented, and using them in downstream analysis can lead to biological misinterpretations (e.g. false fusion events, aberrant alternative splicing, erroneous transcript degradation, etc.).

Such errors manifest as off-subdiagonal elements in our ligation heatmap (left panel in figure below), depicting MAS-seq adapter adjacencies found in each read. The filter command removes the off-subdiagonal reads (right panel), ensuring that only high-quality data with confident and model-consistent segmentations are propagated to downstream analysis.

Command help

$ longbow filter --help
Usage: longbow filter [OPTIONS] INPUT_BAM

  Filter reads by whether they conform to expected segment order.

Options:
  -v, --verbosity LVL    Either CRITICAL, ERROR, WARNING, INFO or DEBUG
  -p, --pbi PATH         BAM .pbi index file
  -o, --out-prefix TEXT  Output file prefix  [required]
  -m, --model TEXT       The model to use for annotation.  If the given value
                         is a pre-configured model name, then that model will
                         be used.  Otherwise, the given value will be treated
                         as a file name and Longbow will attempt to read in
                         the file and create a LibraryModel from it.  Longbow
                         will assume the contents are the configuration of a
                         LibraryModel as per LibraryModel.to_json().
                         [default: mas15]
  -f, --force            Force overwrite of the output files if they exist.
                         [default: False]
  --help                 Show this message and exit.

Example

$ longbow filter -o filtered annotated.bam
[INFO 2021-08-12 10:04:26   filter] Invoked via: longbow filter -o filtered annotated.bam
[INFO 2021-08-12 10:04:26   filter] Using The standard MAS-seq 15 array element model.
[INFO 2021-08-12 10:04:28   filter] Writing reads that conform to the model to: filtered_longbow_filter_passed.bam
[INFO 2021-08-12 10:04:28   filter] Writing reads that do not conform to the model to: filtered_longbow_filter_failed.bam
[INFO 2021-08-12 10:04:28   filter] Filtering according to mas15 model ordered key adapters: A, B, C, D, E, F, G, H, I, J, K, L, M, N, O, P
Progress: 8 read [00:00, 970.54 read/s]
[INFO 2021-08-12 10:04:28   filter] Done. Elapsed time: 2.35s.
[INFO 2021-08-12 10:04:28   filter] Total Reads Processed: 8
[INFO 2021-08-12 10:04:28   filter] # Reads Passing Model Filter: 8 (100.00%)
[INFO 2021-08-12 10:04:28   filter] # Reads Failing Model Filter: 0 (0.00%)
[INFO 2021-08-12 10:04:28   filter] Total # correctly ordered key adapters in passing reads: 110
[INFO 2021-08-12 10:04:28   filter] Total # correctly ordered key adapters in failing reads: 0
[INFO 2021-08-12 10:04:28   filter] Avg # correctly ordered key adapters per passing read: 13.7500 [16]
[INFO 2021-08-12 10:04:28   filter] Avg # correctly ordered key adapters per failing read: 0.0000 [16]