These are presently reported for both junction and segment IES features.

However for junction features, which are zero-length, ta_pointer_start equals start coordinate if the IES is bounded by a TA repeat. This results in coordinate for ta_pointer_start being correct for python zero-based numbering system but not for GFF.

For junction features, ta_pointer_end equals ta_pointer_start, which doesn't make sense either.

Maybe just have a ta_offset attribute that states how many bases the start/end coordinates need to be adjusted. This would then be robust against any change in coordinate systems and also be suitable for both insertions and deletions.

Using a SAM-style region string like ctg1:200:3000

Retention score does not appear to count all the different length variants

Error message:

Traceback (most recent call last):
  File "/ebio/abt2_projects/ag-swart-blepharisma/data/bleties/envs/bleties_env/bin/bleties", line 349, in <module>
    args.func(args)
  File "/ebio/abt2_projects/ag-swart-blepharisma/data/bleties/envs/bleties_env/lib/python3.7/site-packages/bleties/main.py", line 114, in milraa
    args.min_break_coverage, args.min_del_coverage)
  File "/ebio/abt2_projects/ag-swart-blepharisma/data/bleties/envs/bleties_env/lib/python3.7/site-packages/bleties/Milraa.py", line 1425, in reportPutativeIesInsertSubreads
    rname, gffpos, gffpos, consseq, breakpointid)
  File "/ebio/abt2_projects/ag-swart-blepharisma/data/bleties/envs/bleties_env/lib/python3.7/site-packages/bleties/Milraa.py", line 895, in reportAdjustPointers
    self._refgenome[ctg], ins_start, ins_end, consseq, breakpointid)
  File "/ebio/abt2_projects/ag-swart-blepharisma/data/bleties/envs/bleties_env/lib/python3.7/site-packages/bleties/Milraa.py", line 263, in getPointers
    while iesseq[i] == seq[end+i]:
  File "/ebio/abt2_projects/ag-swart-blepharisma/data/bleties/envs/bleties_env/lib/python3.7/site-packages/Bio/Seq.py", line 323, in __getitem__
    return self._data[index]
IndexError: string index out of range

BleTIES v0.1.4
With Contig_9 and Contig_14 in test data

Analogous to IES retention score

Like MILCOR but with transcriptome as reference instead, e.g. for use case where no reference genome is available.

Distinguish introns from IESs by GT/AG boundaries, and 100% "retention" of introns (vs. IESs, if the library is only an enrichment).

However splicing may result in erroneous indel calls; for minimap2 mapper, "splice" mode is not suitable for our needs because it assumes RNAseq reads are being mapped onto genomic reference, not the other way around. What we could do is to compare predicted intron positions on different isoforms, if available.

Improve readability of output files

To get a MAC+IES assembly, e.g. for primer design and annotation

Both MILRET and MILRAA count only reads that completely span an IES and either contain an insert or do not. If inserts/IESs are too long, the mapping will have a clip instead of an insert. For short IESs this is negligible, but as IESs are longer, a greater proportion of mapped IES+ reads will have clips, but these are currently ignored in calculating the retention score.

I tried using Milraa (both released version and development version) on Nanopore reads and the Oxytricha Mac genome from https://knot.math.usf.edu/mds_ies_db/downloads.html. I used the --secondary=no and --MD options as instructed when first aligning using minimap2. However, I get a "String index out of range error" as follows:

Traceback (most recent call last):
File ".../.conda/envs/bleties/bin/bleties", line 386, in
args.func(args)
File ".../.conda/envs/bleties/lib/python3.7/site-packages/bleties/main.py", line 152, in milraa
args.subreads_cons_len_threshold)
File ".../.conda/envs/bleties/lib/python3.7/site-packages/bleties/Milraa.py", line 1486, in reportPutativeIesInsertSubreads
rname, gffpos, gffpos, consseq, breakpointid)
File ".../.conda/envs/bleties/lib/python3.7/site-packages/bleties/Milraa.py", line 917, in reportAdjustPointers
self._refgenome[ctg], ins_start, ins_end, consseq, breakpointid)
File ".../.conda/envs/bleties/lib/python3.7/site-packages/bleties/Milraa.py", line 263, in getPointers
while iesseq[i] == seq[end+i]:
File ".../.conda/envs/bleties/lib/python3.7/site-packages/Bio/SeqRecord.py", line 455, in getitem
return self.seq[index]
File ".../.conda/envs/bleties/lib/python3.7/site-packages/Bio/Seq.py", line 323, in getitem
return self._data[index]
IndexError: string index out of range.

Add user option to define regions that are to be skipped in initial MILRAA step. For example with GFF file of low complexity regions, telomere regions, because these tend to attract poor quality mappings.

Alternative is to filter the MILRAA GFF output afterwards with a mask definition file.

Define specific region, e.g. for testing

Change native output to JSON or TSV, add GFF only as a layer on top. Makes it much easier to troubleshoot coordinate errors!

Describe the bug
Insert module does not handle multi-line GFF3 entries, e.g. a CDS that is split across several lines (separated by introns), but where each line shares the same ID attribute. The Gff class in SharedFunctions has to be updated to handle multi-line Gff entries. This was not considered initially because IES entries are always single-line.

Please attach/copy the following:

NA

• The command that resulted in the error
• The log file from the BleTIES run (default name: bleties.log)
• Any error messages that appeared on the console

Version info (please complete the following information):

• OS: Linux
• BleTIES version 0.1.10-alpha
• How you installed BleTIES: Conda

As of commit 6adc6bf the TA-boundary coordinates are updated to match new insert assembly by BleTIES Insert and reported in the GFF output as attributes. Previously the coordinates were still relative to pre-insert assembly.

This change should be documented.

In MILTEL, clipped segments without telomeres could be aligned to the rest of the assembly

e.g. SPOA, NCRF

Currently they assume that input reads are CCS or error corrected.

ZMW coverage would correspond more closely to physical coverage

It's not clear from the README or program help, whether the normal operation of MILRET should be to provide it with the gff output from MILRAA, or to run it without this - please clarify.

use tempfile.NamedTemporaryFile to handle the input files for aligner

Processing of IES candidates in fuzzy mode to get consensus sequence is slow. Speed up with parallelization?

Implement in MILCOR, correlations would be relative to distance between IESs

Currently, subreads mode only reports inserts relative to reference. CCS mode reports both insertions and deletions relative to the reference.

Running bleties with the MILRAA option gives:

File "..miniconda3/envs/bleties/lib/python3.7/site-packages/bleties/main.py", line 35, in check_env
    dep_vers['muscle'] = re.search(r"v\S+", dep_vers['muscle']).group(0)
AttributeError: 'NoneType' object has no attribute 'group'

Running the command:

muscle -version

muscle 5.1.linux64 []
Built Feb 24 2022 03:16:15

I couldn't see the "junction" annotations when I imported them into Geneious. I could however see the ones imported along with the mapped reads from the BAM file. Can we chose a different name for these annotations (e.g. "IES_junction")? I relabelled my "junction" annotations, and was able to see them...

Here's a nested 18 bp candidate:

Would one get a lot of false positives if one were to chose, e.g. a value of 15?