Abnormal color gradient observed when plotting using the dmr isoform command

[rain-zjg]

Hello,
Modkit team. I used the following command to plot the m⁶A modification levels of transcripts corresponding to specific genes (modkit v.0.6.4).

modkit dmr isoform \
  --gtf ~/genome/gencode.v48.annotation.gtf \
  --single-code "a" \
  --min-valid-coverage 5 \
  --marker-positions 48960898 \
  --gene-name BAX \
  --plot test-A498-BAX \
  ../merge.A498.bed.gz test.A498-BAX.bed

However, I found that the m⁶A sites in the results are almost all blue (In other words, the modification levels are all close to 0). But in my input file, the modification levels range from 0 to 1, so it is unlikely that all of them are below 0.5, right?

I extracted some examples with relatively high modification levels.

#chrom  chromStart      chromEnd        name    score   strand  thickStart      thickEnd        color   valid_coverage  percent_modified count_modified   count_canonical count_other_mod count_delete    count_fail      count_diff      count_nocall
ENST00000502487.5|ENSG00000087088.21|BAX-206    627     628     m       15      +       627     628     255,0,0 15      60.00   9       600       20      0       0
ENST00000502487.5|ENSG00000087088.21|BAX-206    1296    1297    a       55      +       1296    1297    255,0,0 55      98.18   54      100       2       0       0
ENST00000345358.12|ENSG00000087088.21|BAX-202   651     652     a       1521    +       651     652     255,0,0 1521    97.24   1479    411       10      38      3       0
ENST00000513545.5|ENSG00000087088.21|BAX-210    716     717     a       119     +       716     717     255,0,0 119     94.12   112     702       2       1       0
ENST00000293288.12|ENSG00000087088.21|BAX-201   458     459     a       243     +       458     459     255,0,0 243     51.44   125     113       5       5       312     33      0
ENST00000293288.12|ENSG00000087088.21|BAX-201   1214    1215    a       674     +       1214    1215    255,0,0 674     97.63   658     151       3       21      3       0

So what could be causing this issue? If you have time, I would appreciate your advice. Thank you!

Best regards,
Rain
2026/07/27

[ArtRand]

Hello @rain-zjg,

Any chance you could send me the rows from test.A498-BAX.bed corresponding to position 48960898? If you'd rather not post it on GitHub you can email me at art.rand [at] nanoporetech [dot] com

[rain-zjg]

Hello ArtRand, thank you for your quick reply. I can send you the relevant files.

This is the result of the differential modification analysis:
test.A498-BAX.txt

These are the corresponding records from the original bedMethyl file:
test1.BAX.simplified.txt