You can install DEUS from GitHub with:
if (!require("devtools")) install.packages("devtools", repos='http://cran.us.r-project.org') devtools::install_github("timjeske/DEUS", build_opts = c("--no-resave-data", "--no-manual"))
Alternatively, you can install DEUS from a local copy.
First, you need to unzip it:
Then you can install it in R:
if (!require("devtools")) install.packages("devtools", repos='http://cran.us.r-project.org') devtools::install_local("./DEUS-master", build_opts = c("--no-resave-data", "--no-manual"))
The following system libraries are required:
If these system dependencies are fulfilled, all required R packages should be installed automatically.
Further, the DEUS pipeline requires the following tools:
During the annotation step, a BLAST database is used to determine possible biological sources of each unique sequence. Depending on the use-case, two options are available how an appropriate database can be obtained:
update_blastdb.pl, which is part of the BLAST+ suite.
makeblastdband some fasta sequences:
cat DASHR.fa Homo_sapiens.GRCh37.75.cds.all.fa Homo_sapiens.GRCh37.75.ncrna.fa > HumanDB.fa makeblastdb -in HumanDB.fa -dbtype nucl -parse_seqids
In order to execute the differential expression analysis, a file specifying different conditions is necessary. Each condition file has to contain sample names, the respective FASTQ files as well as the matching condition. The file should be in TSV format (Tab-Separated Values).
sampleName sample condition cond1_s1 cond1_s1.fwd.fq.gz cond1 cond1_s2 cond1_s2.fwd.fq.gz cond1 cond2_s1 cond2_s1.fwd.fq.gz cond2 cond2_s2 cond2_s2.fwd.fq.gz cond2
To test your setup, you can run our test script that will perform DEUS on our simulated example data. It can also be used as a template for further analyses. You can find the test script in the DEUS GitHub repository. An additional sample script performing an extended analysis using sequence clusters is also provided in the repository. The individual steps of both scripts are explained in Run DEUS.