The public folder contains:

The reference (along with the annotation) used to map the reads:
GRCm39.genome.fa
GRCm39.genome.fa.fai
gencode.vM31.basic.annotation.gtf

The subfolders in the public folder:

- Ralf_Nuria_standardDRS_heart_m6A
- Ralf_Nuria_shotgun_heart_m6A
- Ralf_Nuria_shotgun_tissues_m6A
- Ralf_Nuria_standardDRS_heart_m6A_merged_good
- Ralf_Nuria_shotgun_heart_m6A_merged_good
- Ralf_Nuria_shotgun_tissues_m6A_merged_good

Basecalling was done with MoP3/mop_preprocess pipeline (with the m6A basecaller) on all of the samples. On these basecalled fast5 files the modPhred pipeline was ran to produce the mod files (the .bams of the individual samples were produced with minimap2 within the modPhred pipeline and they contain the encoded modifications).

Each folder also contains the corresponding fastq files of the samples. Note that the fastq files provided here are the ones originally produced by the mop_preprocess pipeline (which outputs the reads in a single file per sample) that the basecalling was done with (but for the merged cases, they were merged together later on).

Since modPhred needs to receive certain filtering options (in this case: setting the coverage such that only the positions with >=25 in AT LEAST 1 SAMPLE are reported) in order to produce the modifications (mod) file, the pipeline is ran on a particular subset of samples that we wish to compare:

- Ralf_Nuria_standardDRS_heart_m6A:
mettl3 and sham samples

- Ralf_Nuria_shotgun_heart_m6A:
s_mettl3, s_sham and s_tac samples

- Ralf_Nuria_shotgun_tissues_m6A:
s_sham and cb samples
s_sham and cx samples
s_sham and kd samples
s_sham and lg samples
s_sham and lv samples
s_sham and sk samples

Note that for the 6 tissues samples, the s_sham samples from shotgun_heart were ran with the shotgun samples of each other tissue, so a mod file was produced for each one of those cases.

Note that at this level there was still no filtering of the bad-quality samples, so the mod files are composed of all of the samples, including the bad-quality replicates.

However, in the postprocessing of the modPhred results it became obvious that including the low-quality samples was not giving meaningful results, hence the corresponding columns for those samples were edited out of the originally made mod files and these mod_good files were then used in the postprocessing.

In addition, only these "good" samples were used to produce the merged .bams for each condition and in the repeated analysis, since it became evident that for certain tissues the coverage was still not satisfactory across the conditions, for many of the sites:

merged_mettl3:
mettl3_ko2_stripped
mettl3_ko3_stripped

merged_sham:
sham3_stripped
sham4_stripped
sham5_stripped

s_sham:
s_sham1_stripped
s_sham3_stripped
s_sham4_stripped
s_sham5_stripped

merged_s_tac:
s_tac5_stripped
s_tac6_stripped
s_tac7_stripped
s_tac8_stripped

merged_cb:
cb2_stripped
cb3_stripped

merged_cx:
cx1_stripped
cx2_stripped

merged_kd:
kd1_stripped
kd2_stripped
kd3_stripped

merged_lg:
lg1_stripped
lg3_stripped

merged_lv:
lv2_stripped
lv3_stripped

merged_sk:
sk1_stripped
sk3_stripped



IGV visualisation of the modifications:


Open the .bam files, the reference .fa file, the annoteation .gtf file and the modification mod file.


For a particular .bam file, set in the R-click menu:

Shade base by quality
Show mismatched bases
Show all bases

Show Coverage track

Show Alignment track

Collased or Squished (if too many reads)


Alternativelly, these could be set globaly from the main menu: 

"View > Preferences > Alignments"

Show alignment track
Show coverage track

Show mismatched bases
Show all bases

"View > Preferences > Third Gen"

switch off:

Quick consensus mode