Many meme files in the de novo motif step

[linfanxiao]

Hi developers,

I meet an issue with the output of the de novo motif step. I get the endless run like this

[1] "Sampling 2000 (th: 13.9) peaks over 1014329 possible ones."
Writing results to output directory 'motif_detection/meme_EX51VSST51/meme_76'.
Initializing the motif probability tables for 2 to 1997 sites...
nsites = 1980
Done initializing.
SEEDS: highwater mark: seq 1996 pos 22
BALANCE: samples 1997 residues 43934 nodes 60 residues/node 732

seqs=  1997, min=  22, max=   22, total=    43934

motif=1
SEED WIDTHS: 4 6 8 11 14
em: w=   4, psites=   2, iter=  40  
motif=2
SEED WIDTHS: 4 6 8 11 14
em: w=   4, psites=   2, iter=  40  
motif=3
SEED WIDTHS: 4 6 8 11 14
em: w=   4, psites=   2, iter=  40  
motif=4
SEED WIDTHS: 4 6 8 11 14
em: w=   4, psites=   2, iter=  40  
motif=5
SEED WIDTHS: 4 6 8 11 14
em: w=   4, psites=   2, iter=  40  
[2024-06-11 19:32:09] Potentials motifs:.
[2024-06-11 19:32:09] Discovered motifs: CCWGG.
Do you want to continue the research for motifs? (Y/N) 
Y
[2024-06-11 19:32:15] Removing explained peaks.
[1] "Sampling 2000 (th: 13.9) peaks over 1014329 possible ones."
Writing results to output directory 'motif_detection/meme_EX51VSST51/meme_77'.
Initializing the motif probability tables for 2 to 1997 sites...
nsites = 1980

all motif is CCWGG
when i use parameters "-a" only output a file named "MEME_1"

This is my command:
singularity run --no-home -w nd_example --writable

nanodisco motif -p 60 -b EX51VSST51 -d EX51VSST51_merged/EX51VSST51_difference.RDS -o EX51VSST51_motif -r ncbi_dataset/data/GCF_000401755.1/GCF_000401755.1_Escherichia_coli_ATCC_25922_genomic.fasta

Do I only need the first meme file?

[touala]

Hello @linfanxiao,

It looks like the only motif found was CCWGG and all rerun on the remaining peaks are not able to find any new motifs. You could try with -a and select different parameters for selecting number of peaks (-t and --nb_peaks). It also possible that this strains only contains one methylated motif.

Best,

Alan