Using STARlong for Oxford Nanopore DRS reads....

[Nick Schurch]

Has anyone tried and had success with this?

I've tried it with start2.5 and the recommended setting for PacBio Isoseq data, but with very marginal success. These are the options I use:

--runMode alignReads
--readNameSeparator space
--outFilterMultimapScoreRange 1
--outFilterMismatchNmax 2000
--scoreGapNoncan -20
--scoreGapGCAG -4
--scoreGapATAC -8
--scoreDelOpen -1
--scoreDelBase -1
--scoreInsOpen -1
--scoreInsBase -1
--alignEndsType Local
--seedSearchStartLmax 50
--seedPerReadNmax 100000
--seedPerWindowNmax 1000
--alignTranscriptsPerReadNmax 10000
 --alignTranscriptsPerWindowNmax 10000
--genomeDir STAR_2.5_sjdbOverhang75

The result is 

> more Log.final.out
                                 Started job on | Aug 01 17:22:47
                             Started mapping on | Aug 01 17:23:32
                                    Finished on | Aug 01 21:02:18
       Mapping speed, Million of reads per hour | 0.03


                          Number of input reads | 116617
                      Average input read length | 900
                                    UNIQUE READS:
                   Uniquely mapped reads number | 532
                        Uniquely mapped reads % | 0.46%
                          Average mapped length | 19.71
                       Number of splices: Total | 194
            Number of splices: Annotated (sjdb) | 34
                       Number of splices: GT/AG | 193
                       Number of splices: GC/AG | 1
                       Number of splices: AT/AC | 0
               Number of splices: Non-canonical | 0
                      Mismatch rate per base, % | 3.73%
                         Deletion rate per base | 0.40%
                        Deletion average length | 1.91
                        Insertion rate per base | 0.10%
                       Insertion average length | 2.00
                             MULTI-MAPPING READS:
        Number of reads mapped to multiple loci | 1226
             % of reads mapped to multiple loci | 1.05%
        Number of reads mapped to too many loci | 71
             % of reads mapped to too many loci | 0.06%
                                  UNMAPPED READS:
       % of reads unmapped: too many mismatches | 0.00%
                 % of reads unmapped: too short | 93.84%
                     % of reads unmapped: other | 4.59%
                                  CHIMERIC READS:
                       Number of chimeric reads | 0
                            % of chimeric reads | 0.00%

Clearly it is trying to map the reads but it seems it can't find good seeds to begin with. I'm trying again with:

--seedSearchStartLmax 20

but also wondering if my genome index isn't helping. I'm using an index build for 75bp Illumina reads...

Thoughts?

[Alexander Dobin]

Hi Nick,

STAR will likely not work well for reads with a large rate of indels.

The parameters you may want to tweak aggressively are

--seedSearchStartLmax  - reduce to 10 or even below

--winAnchorMultimapNmax - increase to 1000 or more

--seedMultimapNmax - increase to 100000 or more

If you have a few reads that you can share, I can play with parameters to see what is the best the current algorithm can do.

Cheers

Alex

[Nick Schurch]

Thanks Alex, I'll give that a try. I'll check with my PI and make sure he's happy before I do splashing the reads around ;)

[Nick Schurch]

I tried these parameters but it resulted in a Fatal Error during the alignment:

EXITING because of FATAL error: too many pieces pere read
SOLUTION: increase input parameter --seedPerReadNmax
Aug 08 18:09:03 ...... FATAL ERROR, exiting

Seems like it really doesn't like --seedSearchStartLmax 10. I'll try it with 20 and see if that goes through...

. 

[Alexander Dobin]

Hi Nick,

it seems you would have to further increase --seedPerReadNmax until this error disappears.

Cheers

Alex