search

uschwartz / nucMACC

Pipeline to call hyper and hypo accessible nucleosomes and nucleosomes with non-canonical structure based on differential MNase-seq data
MIT License
3 stars 1 forks source link

The following chromosome names did not match between the bigwig files #12

Closed davhum closed 1 month ago

davhum commented 1 month ago

I am trying this pipeline against hg38 genome. The pipeline gets most of the way through but then falls over with the following message:

The following chromosome names did not match between the bigwig files chromosome length chrUn_KI270420v1 2321 chrUn_KI270744v1 168472 chrUn_KI270583v1 1400 chrUn_KI270363v1 1803 chr1_KI270714v1_random 41717 chrUn_KI270391v1 1484 chr22_KI270737v1_random 103838 chr15_KI270727v1_random 448248 chrUn_KI270518v1 2186 chrUn_KI270422v1 1445 chr9_KI270720v1_random 39050 chrUn_KI270382v1 4215 chr1_KI270708v1_random 127682 chrUn_KI270465v1 1774

This message I believe comes from deeptools. I wonder if this is a warning message from deeptools or an actual error. Should I be aligning to a hg38 genome with only full chr entries?

uschwartz commented 1 month ago

I would recommend using genome assemblies without haplotype/patch regions such as the Ensembl primary assembly files. Haplotype/patch regions are prone to produce artifacts and I am not sure that they prodvide reliable information about nucleosome organisation.

Not sure when exactly (which process?) this message comes and I would need further information to figure out what happend exactly. The pipeline automatically produces the chromosome length information basically out of the bowtie index. It might happen that no reads align to those patches and they are lost in the bigwig files and deepltools will complain about that, but this should not run in an error and abort the run.

davhum commented 1 month ago

Thanks - you were spot on. I made a new bowtie2 index and was able to get past this point. I unfortunately have another error which I will post in another thread.

The error reported in this thread was caused when running the following deeptools command:

computeMatrix reference-point -S