Reorganize repo. Assume use of zibra/zibra Docker image.

.gitignore

@@ -1,7 +1,5 @@
 # For zika-seq repo #
 environment*
-data/libraries/
-scripts/ssw_lib.py
 
 # Jekyll #
 ##########

README.md

@@ -1,55 +1,68 @@
 # Experimental protocols and bioinformatic pipelines for Zika genome sequencing
 
+#### Allison Black<sup>1,2</sup>, Barney Potter<sup>2</sup>, Nicholas J. Loman<sup>3</sup>, Trevor Bedford<sup>2</sup>
+
+<sup>1</sup>Department of Epidemiology, University of Washington, Seattle, WA, USA, <sup>2</sup>Vaccine and Infectious Disease Division, Fred Hutchinson Cancer Research Center, Seattle, WA, USA, <sup>3</sup>Institute of Microbiology and Infection, University of Birmingham, Birmingham, UK
+
 ## Install
 
-Clone the repo:
+Clone the repo and load submodules:
 
     git clone https://github.com/blab/zika-seq.git
+    git submodule update --init --recursive
 
-Install Python dependencies:
+## Data sync
 
-    pip install -r requirements.txt
+Primary sequencing data lives on the Rhino FHCRC cluster at:
 
-Install [SSW Library](https://github.com/mengyao/Complete-Striped-Smith-Waterman-Library):
+    /fh/fast/bedford_t/data/
 
-    git clone https://github.com/mengyao/Complete-Striped-Smith-Waterman-Library.git
-    cd Complete-Striped-Smith-Waterman-Library/src/
-    cp libssw.so <path to zika-seq>/zika-seq/scripts/
-    cp ssw_lib.py <path to zika-seq>/zika-seq/scripts/
+And locally on Meristem drive at:
 
-Install [marginAlign](https://github.com/benedictpaten/marginAlign):
+    /Volumes/Meristem/data/
 
-    git clone https://github.com/benedictpaten/marginAlign.git
-    cd marginAlign
-    git submodule update --init --recursive
-    make
-    export PATH=<path to marginAlign>/marginAlign/:$PATH
+To sync Meristem to Rhino, run:
 
-Install [samtools](https://github.com/samtools/samtools):
+    rsync -azP tbedford@rhino.fhcrc.org:/fh/fast/bedford_t/data/ /Volumes/Meristem/data/
 
-    brew tap homebrew/science
-    brew install samtools
+Replacing `tbedford` with your username.
 
-## Data sync
+This `data/` directory is assumed to follow [a particular schema](https://github.com/blab/zika-seq/blob/master/data-schema.md).
 
-From `zika-seq` run:
+## Bioinformatic pipeline
 
-    rsync -azP tbedford@rhino.fhcrc.org:/fh/fast/bedford_t/zika-seq/data/ data/
+Here, we use the ZiBRA project bioinformatic pipeline at [zibraproject/zika-pipeline](https://github.com/zibraproject/zika-pipeline/). This pipeline is instantiated in the Docker image [zibra/zibra](https://hub.docker.com/r/zibra/zibra/). Data processing is done using Docker.
 
-Replacing `tbedford` with your username.
+### Data volume
 
-## Bioinformatic pipeline
+Create a named data volume that mirrors local `data/` to `data/` within container:
+
+    docker create --name zibra-data -v /Volumes/Meristem/data:/data zibra/zibra    
+
+This is to get data into the Docker container. Note that the path to local directory has to be an absolute path.
+
+Create a named data volume for a single sample:
+
+    docker create --name zibra-data-lb01-nb01 -v /Volumes/Meristem/data/usvi-library1-2016-12-10/basecalled_reads/pass_demultiplex/NB01:/data zibra/zibra
+
+### Build volume
+
+Create a named data volume that mirrors local `build/` to `build/` within container:
+
+    docker create --name zibra-build -v /Volumes/Meristem/build:/build zibra/zibra
+
+This is to get data out of the Docker container. Note that the path to local directory has to be an absolute path.
 
-Data lives in the [`data/`](data/) directory and is not versioned within the repo. Directory structure described in its [README.md](data/).
+### Start
 
-### Base calling
+Enter docker image:
 
-Convert raw MinION output to FAST5
+    docker run -t -i --volumes-from zibra-data --volumes-from zibra-build zibra/zibra /bin/bash
 
-    metrichor-cli -a <API KEY> -w 1289 -f - -i <directory_with_fast5s> -o downloads
+Run single sample script within image:
 
-### Run pipeline
+    ./scripts/go_single_sample_r94.sh refs/KJ776791.2.fasta NB03 metadata/v2_500.amplicons.ver2.bed
 
-Run poretools, marginAlign, samtools:
+## Results
 
-    python run.py
+  * [Initial coverage results from first library are here](depth-coverage/)

data-schema.md

@@ -0,0 +1,20 @@
+# Data schema
+
+## Nanopore reads
+
+Input data to the Zika pipeline arrives in the `data/` directory. This should be [mounted to `data/` in the Docker container](https://github.com/blab/zika-seq#data-volume).
+
+  - `data`
+    - `usvi-library1-2016-12-10` - library
+      - `raw_reads` - squiggle graphs in fast5 format
+        - `pass` - contains `.fast5` files
+      - `basecalled_reads` - basecalled with Metrichor
+        - `pass_demultiplex` - demultiplexed basecalled reads
+          - `NB01` - contains `.fast5` files for NB01 barcode
+          - `NB02` - contains `.fast5` files for NB02 barcode
+          - etc...
+        - `nonNB_demultiplexed` - demultiplexed basecalled reads
+          - `BC01` - contains `.fast5` files for BC01 barcode
+          - `BC02` - contains `.fast5` files for BC02 barcode
+          - etc...        
+        - `fail` - contains `.fast5` files that weren't demultiplexed