Per Unneberg

This will be tricky for the function generic_target_generator. Affected rules are snakemakelib/rules/bio/ngs/qc/rseqc/rseqc_qc_8_summary.rule snakemakelib/rules/bio/ngs/qc/cutadapt/cutadapt_qc_summary.rule snakemakelib/rules/bio/ngs/qc/picard/picard_merge_sam.rule snakemakelib/rules/bio/ngs/qc/picard/picard_do_summary.rule snakemakelib/rules/bio/ngs/qc/qualimap/qualimap_collect_results.rule snakemakelib/rules/bio/ngs/rnaseq/rsem/rsem_summarize_expression_data.rule snakemakelib/rules/bio/ngs/align/star/star_summarize_alignment_runs.rule

Missing a ";" in the annotation file causes STAR to misname references...

#41 was fixed, not #40, in a3b3e2a

Even better: allow negative indexing, as in samples=[-"sample1", "sample2"]

Chiming in here. We noted the issue of contigs not being sorted according to the input order of the assembly file. The culprit here is not biopython, but rather BCBio.GFF....

@hyanwong As requested I'm following up on the discussion in #2026, where I was asking about the function [local_gnn](https://github.com/mcveanlab/treeseq-inference/blob/master/src/analyse_human_data.py#L148). I did go through the code, and now that I've reread...

I have a question related to the interpretation of locations and coordinates. I'm currently using the documentation example in slendr to simulate a test dataset for a visualization of GNNs...

Do we always want to enforce large index?

Actually, running `bowtie2-build-l` gives the following message: ``` *** Warning *** 'bowtie2-build-l' was run directly. It is recommended that you run the wrapper script 'bowtie2-build' instead. ``` One should therefore...

Alternatively one could have a target that specifically lists the krakenuniq output, e.g., ``` rule krakenuniq: input: all_input["krakenuniq.krona"] ```