Merge pull request #7 from Russel88/dev

0.1.18

Author: trinezac

README.md

@@ -58,17 +58,18 @@ maginator ... --cluster qsub --cluster_info "-l nodes=1:ppn={cores}:thinnode,mem
 
 ## Test data
 
-A test set can be found in the test_data directory. 
+A test set can be found in the maginator/test_data directory. 
 1. Download the 3 samples used for the test at SRA: https://www.ncbi.nlm.nih.gov/sra?LinkName=bioproject_sra_all&from_uid=715601 with the ID's dfc99c_A, f9d84e_A and 221641_A
-2. Change the paths to the read-files in reads.csv
-3. Unzip the contigs.fasta.gz 
-4. Run MAGinator
+2. Clone repo: git clone https://github.com/Russel88/MAGinator.git
+3. Change the paths to the read-files in reads.csv
+4. Unzip the contigs.fasta.gz 
+5. Run MAGinator
 
 MAGinator has been run on the test data on a slurm server with the following command:
-```
+```sh
 maginator --vamb_clusters clusters.tsv --reads reads.csv --contigs contigs.fasta --gtdb_db data/release207_v2/ --output test_out --cluster slurm --cluster_info "-n {cores} --mem {mem_gb}gb -t {runtime}" --max_mem 180
 ```
-The expected output can be found in test_data/test_out (excluding the GTDB-tk folders, phylogeny alignments and BAM-files due to size limitations)
+The expected output can be found as a zipped file on Zenodo: https://doi.org/10.5281/zenodo.8279036
 
 ## Recommended workflow 
 
@@ -88,14 +89,23 @@ sed 's/@/_/g' vamb/clusters.tsv > clusters.tsv
 
 Now you are ready to run MAGinator.
 
+## Functional Annotation
+
 To generate the functional annotation of the genes we recommend using EggNOG mapper (https://github.com/eggnogdb/eggnog-mapper).
 
 You can download it and try to run it on the test data
-```
+```sh
 mkdir test_out/functional_annotation
 emapper.py -i test/genes/all_genes_rep_seq.fasta --output test_out/functional_annotation -m diamond --cpu 38
 ```
 
+The eggNOG output can be merged with clusters.tsv and further processed to obtain functional annotations of the MAG, cluster or sample levels with the following command:
+```sh
+(echo -e '#sample\tMAG_cluster\tMAG\tfunction'; join -1 1 -2 1 <(awk '{print $2 "\t" $1}' clusters.tsv | sort) <(tail -n +6 annotations.tsv | head -n -3 | cut -f1,15 | grep -v '\-$' | sed 's/_[[:digit:]]\+\t/\t/' | sed 's/,/\n/g' | perl -lane '{$q = $F[0] if $#F > 0; unshift(@F, $q) if $#F == 0}; print "$F[0]\t$F[1]"' | sed 's/\tko:/\t/' | sort) | awk '{print $2 "\t" $2 "\t" $3}' | sed 's/_/\t/' | sort -k1,1 -k2,2n) > MAGfunctions.tsv
+```
+In this case the KEGG ortholog column 15 was picked from the eggNOG-mapper output. But by cutting e.g. column number 13, one would obtain GO terms instead. Refer to the header of the eggNOG-mapper output for other available functional annotations e.g. KEGG pathways, Pfam, CAZy, COGs, etc.
+
+
 ## MAGinator workflow
 
 This is what MAGinator does with your input (if you want to see all parameters run maginator --help):

conda_build/meta.yaml

@@ -0,0 +1,5 @@
+name: checkm-genome
+channels:
+  - bioconda
+dependencies:
+  - checkm-genome

maginator/recommended_workflow/envs/checkm.yaml

@@ -0,0 +1,4 @@
+library(BSgenome.Hsapiens.UCSC.hg19.masked)
+genome <- BSgenome.Hsapiens.UCSC.hg19
+out_file <- file.path(snakemake@output[["hg19"]])
+export(genome, out_file)

maginator/recommended_workflow/envs/import_hg_19.R

@@ -0,0 +1,5 @@
+name: metabat2
+channels:
+  - bioconda/label/cf201901
+dependencies:
+  - metabat2

maginator/recommended_workflow/envs/metabat2.yaml

@@ -0,0 +1,13 @@
+channels:
+  - bioconda
+  - conda-forge
+  - r
+dependencies:
+  - biopython=1.79
+  - pandas=1.4
+  - bbmap=38.96
+  - sickle-trim=1.33
+  - spades=3.15.5 
+  - samtools=1.10
+  - bwa-mem2=2.2.1
+  - bioconductor-bsgenome.hsapiens.ucsc.hg19.masked=1.3.993

maginator/recommended_workflow/envs/preprocess.yaml

@@ -0,0 +1,5 @@
+name: samtools
+channels:
+  - bioconda
+dependencies:
+  - samtools

maginator/recommended_workflow/envs/samtools.yaml

@@ -0,0 +1,14 @@
+name: vamb
+channels:
+  - pytorch
+  - conda-forge
+  - bioconda
+dependencies:
+  - pytorch
+  - pip
+  - torchvision
+  - cudatoolkit=10.2
+  - pysam
+  - numpy=1.20
+  - pip:
+     - git+https://github.com/RasmussenLab/vamb@v3.0.8

maginator/recommended_workflow/envs/vamb.yaml

@@ -1,6 +1,6 @@
 channels:
- - bioconda
  - conda-forge
+ - bioconda
  - biobuilds
 dependencies:
  - biopython=1.79

maginator/workflow/envs/phylo.yaml

@@ -1,4 +1,8 @@
+# New version
+## 1) Update version in setup.py and commit and push
+## 2) Pull request of dev into main
+## 3) Make release on GitHub
+## 4) Run this code:
 rm -r maginator.egg-info/ dist/ build/
 python setup.py sdist
-python setup.py install
 twine upload dist/*