New dopseq version needed

[amakunin]

Update 20180516 - rewrite as snakemake to improve file & job handling.

Features requested:

• [x] store files by filetype, not by sample. Deduplicate files from same reads, but still use sample name prefixes (symlinks for duplicates?).
• [ ] manual for dopseq output parsing
• [ ] manual for updating existing installation
• [ ] manual for dopseq parameter setting (unify these three as wiki/pdf)
• [x] add sample prefix to repeatexplorer fasta sequence names
• [x] replace bedtools with pybedtools
• [ ] return alternate fasta generation to variation_pipeline
• [ ] input parameters as xls
• [ ] remove sizes dependency
• [ ] mapping stats using pybamstats
• [ ] smart incomplete file detection (resulting from interrupts) (tmp file)
• [ ] statistics in one file for all samples (xls/tsv)
• [x] revisit and test contam_filter algorithm
• [ ] multiple contamination genomes (blast or centrifuge)

Bug fixes needed:

• [x] incorrect sample prefix in stats file
• [ ] incorrect reference genomes in multisample setting
• [ ] random sample running order
• [ ] region script sometimes fails at mapply gsub
• [ ] region script pd end to start
• [ ] region script pdf: chromosomes not drawn entirely if reads not present
• [ ] variation pipeline errors in counting covered bases

[amakunin]

Closed via #15