jordiabante / cpelasm.jl Goto Github PK

I'm very confused about the origin of the two bam files in the toyExample. Based on the bismark_reports, the bam came from aligning PE reads to two haplotypes separately. I assume there should be a process to divide reads into two groups (correspond to two haplotypes), but two bams have the same aligned reads (396 reads).

For example, the read "EASX:X:X:X:X:33:33_1:Y:0:NNNNNN" appear in two bam files. Is it correct? Any read should only have one origin.

Can you share how to create two bam files? thanks

Dear CpelAsm Team,
I have followed your instructions to calculate the results, but they seem to lack statistical significance, and the p-values are close to 1. I cannot identify where the issue might be. Do you know what could be causing this situation? I really hope to receive your assistance.

I'm getting the error "Found overlap between homozygous and heterozygous CpG sites. Check FASTA file is masked" when I try to use CpelAsm, and I'm trying to figure out which part of my input is causing it. I looked at the relevant part of the source code, but I'm new to Julia and wasn't able to glean much.

What exactly is the overlap referring to; is it a feature of the VCF file, the BAMs, or an interaction of them?

My input files are the N-masked genome fasta and three BAMs (genome1, genome2, unassigned) created by SNPsplit plus a VCF file. The VCF is derived from the file that I used to N-mask the genome and prepare the SNP list for SNPsplit, but I modified it to follow the VCF in the example (e.g., combining two 1/1 and 0/0 samples into one 1/0 sample and adding the haplotype tag).

Thank you for your time.

Dear CpelAsm Team,

In the original text, you mentioned: "Note, however, that the clear superiority of the CPEL method over existing methods comes at a substantial computational cost (it took about 48 hours using 20 CPUs to process one WGBS sample in the real data), which necessitates the use of parallel processing using a computer cluster." However, when I run CpelAsm, it appears to use only one CPU. Due to my limited familiarity with the Julia language, could you please guide me on how to enable parallel processing? Thank you very much.

Dear CpelASM Team,

I really would like to run your software on my datasets, but unfortunately, it does not run under the latest Julia version on my device. Moreover, I can not find Julia in the version 1.3.0 that is recommended in the ReadMe. Could you give me a hint with which version of Julia I may have success or what I might overlooked?

Thanks in advance, Christiane