probability_modified column is empty

[BioinfoHub-PeiQinNg]

Hi there,

I have tried running m6anet run-inference to predict m6A modifications on my nanopore data.

This was done based on the quick start tutorial https://m6anet.readthedocs.io/en/latest/quickstart.html

However, when I have a look at the output file, instead of indicating a 0 or a 1 as shown in the examples, mine turns out to be a blank column.

For your reference I have attached a snippet of my output file:

transcript_id, transcript_position, n_reads, probability_modified, kmer
2L, 74989, 31, ,TGACT
2L, 75008, 32, ,AAACT
2L, 75100, 33, ,TAACA
2L, 75106, 31, ,TGACT
2L, 75151, 35, ,AAACT

If you can provide me some guidance with this that will be great. Thank you!

[Stakaitis]

Your n_reads column is also empty which is probably the initial problem.
Could you show the head of m6anet-dataprep outputs:

• data.log
• data.readcount
• eventalign.index
• data.index

[BioinfoHub-PeiQinNg]

Hi there,
Thank you for getting back to me.

These are what the few output head of m6anet-dataprep looks like

• head data.log
  2L: Data preparation ... Done.
  2R: Data preparation ... Done.
  3L: Data preparation ... Done.
  3R: Data preparation ... Done.
  4: Data preparation ... Done.
  X: Data preparation ... Done.
  Y: Data preparation ... Done.
  dmel_mitochondrion_genome: Data preparation ... Done.
  Unmapped_Scaffold_8: Data preparation ... Done.
  3Cen_mapped_Scaffold_31: Data preparation ... Done.

• head data.readcount
  transcript_id,transcript_position,n_reads
  2L,49675,1
  2L,51533,1
  2L,51692,1
  2L,51748,1
  2L,51783,1
  2L,51792,1
  2L,51883,2
  2L,51927,1
  2L,52054,2

• eventalign.index
  transcript_id,read_index,pos_start,pos_end
  2L,0,154,56988
  2L,1,56988,117016
  2L,2,117016,229586
  2L,3,229586,470294
  2L,5,470294,934066
  2L,6,934066,1334426
  2L,7,1334426,1873408
  2L,8,1873408,2155664
  2L,9,2155664,2304450

• head data.index
  transcript_id,transcript_position,start,end
  2L,49675,0,69
  2L,51533,69,138
  2L,51692,138,207
  2L,51748,207,276
  2L,51783,276,345
  2L,51792,345,414
  2L,51883,414,521
  2L,51927,521,590
  2L,52054,590,697

Thank you again for assisting me with this.

[BioinfoHub-PeiQinNg]

I have also provided the command lines that I have used for nanopore eventalign and m6a-dataprep for your reference:

nanopolish eventalign \ --reads {input.fasta} \ --bam {input.bam} \ --genome {input.ref} \ --scale-events > {SAMPLE}-ref.eventalign.txt

m6anet-dataprep --eventalign {input.txt} \ --out_dir {output} \

[chrishendra93]

Thank you @Stakaitis for helping to respond. @BioinfoHub-PeiQinNg your nanopolish eventalign is missing --signal-index argument which tells m6Anet the segment length information

[BioinfoHub-PeiQinNg]

I see. I will give it another go. Thank you so much!

[chrishendra93]

hi @BioinfoHub-PeiQinNg, how is it working with you? Are you able to get the results now?

[BioinfoHub-PeiQinNg]

Hi @chrishendra93 , sorry for the late reply. I manage to get it running now! Thank you so much for your help! Cheers ^^