updated README

NGDSG · Jan 22, 2018 · d2318eb · d2318eb
1 parent 770c840
commit d2318eb
Showing 1 changed file with 61 additions and 54 deletions.
diff --git a/README.md b/README.md
@@ -33,46 +33,52 @@ WORKER_NODES denote the hostnames of the machines running filtering workers
 
 It is assumed the resources are configured so that the WORKER_NODES can can resolve the hostname of MASTER_NODE and STORAGE_NODE and reach them via a network.
 
-BAMSI should work on any cloud provider supporting Ubuntu VMs. You will need Python, and the following packages:
-
-* java
-* git
-* pip
-* rabbitmq-server(only on MASTER_NODE)
-* SAMtools
+BAMSI should work on any cloud provider supporting Ubuntu VMs.
 
+### Setup and configuration ####
 
+Examples will be shown for Ubuntu using apt.
 
-```
-sudo apt-get update
-sudo apt-get install default-jre
-sudo apt-get install git
-sudo apt-get install python-pip
-sudo apt-get install python-dev
-sudo apt-get install zlib1g-dev
-sudo apt-get install samtools
-```
+1. Initial requirements:
 
+* python 2
+* java
+* git
+* pip
 
+     ```
+    $ sudo apt-get update
+    $ sudo apt-get install default-jre
+    $ sudo apt-get install git
+    $ sudo apt-get install python-pip
+    $ sudo apt-get install python-dev
+    $ sudo apt-get install zlib1g-dev
+    ```
 
 
-### Setup and configuration ####
+2. Clone the BAMSI git repository
 
-1. Clone the BAMSI git repository
      ```
     $ git clone https://github.com/NGDSG/BAMSI.git
     ```
-2. Change the directory, and install python packages with pip
+
+3. Change the directory, and install python packages with pip
+
     ```
     $ cd BAMSI
-    $ sudo apt-get install python-pip
     $ sudo pip install -r requirements.txt
     ```
-3. Install and start the Celery broker - RabbitMQ (Only on MASTER_NODE)
+
+
+4. On all WORKER_NODES: install SAMtools version 1.x
+
+    See http://www.htslib.org/download/ for download instructions.
+
+
+5. Only on MASTER_NODE: install and start the Celery broker - RabbitMQ
 
    (See https://www.rabbitmq.com/download.html for instructions)
 
-    Example (on ubuntu using apt)
      ```
     $ sudo apt-get install rabbitmq-server
      ```
@@ -89,50 +95,50 @@ sudo apt-get install samtools
 
 
 
-4. On the MASTER_NODE and all the WORKER_NODES: modify the configuration file config.cfg according to your setup:
-
- Example configuration file:
+6. On the MASTER_NODE and all the WORKER_NODES: modify the configuration file config.cfg according to your setup:
 
-```
-[bamsi]
-CELERY_BROKER_URL = amqp://celery:stalk@MASTER_NODE/celery-host
-CELERY_RESULT_BACKEND = amqp
-DATA_PATH=ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/phase3/data/{individual}/alignment/{filename}
-MASTER_IP=MASTER_NODE
-MASTER_PORT=8888
+    Example configuration file:
 
-[storage]
-WEBHDFS_IP=STORAGE_NODE
-WEBHDFS_PORT=50070
-WEBHDFS_PUBLIC_IP=STORAGE_NODE
-WEBHDFS_PUBLIC_PORT=14000
-WEBHDFS_USER=ubuntu
-RESULTS_PATH=/filtered/
-```
+    ```
+    [bamsi]
+    CELERY_BROKER_URL = amqp://celery:stalk@MASTER_NODE/celery-host
+    CELERY_RESULT_BACKEND = amqp
+    DATA_PATH=ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/phase3/data/{individual}/alignment/{filename}
+    MASTER_IP=MASTER_NODE
+    MASTER_PORT=8888
+
+    [storage]
+    WEBHDFS_IP=STORAGE_NODE
+    WEBHDFS_PORT=50070
+    WEBHDFS_PUBLIC_IP=STORAGE_NODE
+    WEBHDFS_PUBLIC_PORT=14000
+    WEBHDFS_USER=ubuntu
+    RESULTS_PATH=/filtered/
+    ```
 
-The settings in the section 'bamsi' are mandatory. They define the location and protocol of the celery host, and define the location from which
- each WORKER_NODE streams the data. To use the public Amazon S3 mirror, use:
+    The settings in the section 'bamsi' are mandatory. They define the location and protocol of the celery host, and define the location from which
+     each WORKER_NODE streams the data. To use the public Amazon S3 mirror, use:
 
-```
-DATA_PATH=http://s3.amazonaws.com/1000genomes/phase3/data/{individual}/alignment/{filename}
-```
+    ```
+    DATA_PATH=http://s3.amazonaws.com/1000genomes/phase3/data/{individual}/alignment/{filename}
+    ```
 
 
 
-The settings in the section 'storage' relate to the storage repository and can be changed accordingly when adding support for other systems.
-Adding a different type of storage repository requires writing a class that implements the interface defined by the class StorageRepositoryBase in tapp.py.
-If no other functionality is required, creating such a class and changing the line
+    The settings in the section 'storage' relate to the storage repository and can be changed accordingly when adding support for other systems.
+    Adding a different type of storage repository requires writing a class that implements the interface defined by the class StorageRepositoryBase in tapp.py.
+    If no other functionality is required, creating such a class and changing the line
 
-    StorageRepository = HDFS()
+        StorageRepository = HDFS()
 
-in tapp.py to initialize your new class instead, and modifying the config file accordingly, should suffice.
+    in tapp.py to initialize your new class instead, and modifying the config file accordingly, should suffice.
 
 
 
-For the current HDFS implementation it is assumed that the namenode is listening on STORAGE_NODE:50070 and
-that a HttpFS server is listening on STORAGE_NODE:14000. These may have to be modified depending on your setup of HDFS.
-(E.g. if the WORKER_NODE is on the same network as the HDFS datanodes, then they can push results to the storage via the namenode on port 50070.
-Otherwise, they should interact with HDFS via the HttpFS server, and WEBHDFS_PORT should also be 14000.)
+    For the current HDFS implementation it is assumed that the namenode is listening on STORAGE_NODE:50070 and
+    that a HttpFS server is listening on STORAGE_NODE:14000. These may have to be modified depending on your setup of HDFS.
+    (E.g. if the WORKER_NODE is on the same network as the HDFS datanodes, then they can push results to the storage via the namenode on port 50070.
+    Otherwise, they should interact with HDFS via the HttpFS server, and WEBHDFS_PORT should also be 14000.)
 
 
 ### Starting the application ####
@@ -167,3 +173,4 @@ There is also a Python API (https://github.com/NGDSG/BAMSI-API) that is possible
 
 
 
+contact: kristiina.ausmees@it.uu.se