Link to this post | posted 16 Feb, 2017 17:48
cdshaffer	The server is reporting service unavailable for me, so its not anything at your end. Unfortunately, I don't think there is anything to do until the server comes back up, since all the ways to install that I know of require running the auto-updater, which is also failing for me.

Link to this post | posted 14 Feb, 2017 17:50

cdshaffer

I realized that even though I don't have time now to write code to create the kind of specific starterator text when doing whole phage reports, that something is better than nothing. So, I just cloned the code I used for creating the text for the individual pham reports into the code for whole phage. I ran a test with phage Amgine and you can download the results of that full page report here. This solution is not ideal as the new individual pham reports give more details on each and every gene in the pham, so individual pham reports on large phams can get quite large. Now, when you put many of those together to make a whole phage report the file can get really really big. For example the whole phage Amgine report is over 1100 pages long. However, this may work better that going to the numerous pham reports on the web. I would suggest using a PDF viewer with good searching tools to quickly find things. If anyone wants the updated code you just need to pull and run the most current version of the byphagewithbase branch from my github repo.

Link to this post | posted 13 Feb, 2017 17:05

cdshaffer

Aaron, If you are doing whole phage then yes that is the current output for that branch. That branch is stuck in the middle of being updated, I have removed the old code for whole phage reports but have not had time to put in the new code to create the output similar to the pre-computed pham reports but with extra text based on which whole phage is being analyzed. Welcome to the cutting edge where you often get cut. I did create a "whole phage" report manually by extracting all the pham numbers for the phage using command line mysql, creating all the pham pdf's using command line starterator, and finally concatenating all the pdf's together using command line ghostscipt. This will create a single PDF with all the pham reports for that phage but is still missing the first few header pages with the map and the list of suggested starts. If you want more details on exact commands let me know. I like your idea of adding the track No. to the text on the track. I have added this as an issue on the starterator github page here: https://github.com/SEA-PHAGES/starterator/issues/26 Not sure when I will have time to get back to coding on starterator, hopefully some over spring break. Edited 13 Feb, 2017 17:50

Link to this post | posted 10 Feb, 2017 16:51

cdshaffer

The answer to your question depends on which version of starterator you want. There is currently version 1.1 which has many bug fixes (not all bugs are fixed just some) and other updates which I did mostly over the summer, it is the master branch at https://github.com/SEA-PHAGES/starterator.git Based on discussions with Welkin, Deb and extra feedback from the in silico workshop, I was able over winter to update a lot of the text output. I am using this code to run the pre-computed pdfs but that version has not been fully tested to make sure it is "release ready", (i.e. it runs on my machine with lots of extra modifications but I have not tested it with the default SEA 2017 VM). The version that is doing all the pre-computing is byphagewithbase branch in my personal repo at: https://github.com/cdshaffer/starterator.git Having said that anyone can install and run any version of the code but it takes a good understanding of unix administration and git: 1. Ensure you have the dependencies (if you are not using the SEA 2017 VM): sudo apt-get install python-pip ncbi-blast+ git sudo pip install PyPDF2 sudo pip install beautifulsoup4 sudo pip install requests 2. Remove all old Starterator files: cd $HOME/Applications rm -rfv $HOME/.starterator rm -rfv $HOME/Applications/Starterator 3. Create new Starterator folder: mkdir $HOME/Applications/Starterator cd $HOME/Applications/Starterator 4. Clone the version you want: git clone https:<insert URL here for the repo you want to clone> 5. Checkout the branch you want: cd $HOME/Applications/Starterator/starterator git checkout <insert branch name here> 6. Run your checked out version of Starterater: bash starterator.sh Please remember, no guarantees for anything in my repo to work on the 2017 VM. However if you try and it does work please do let me know. Edited 13 Feb, 2017 17:04

Link to this post | posted 10 Feb, 2017 16:21

cdshaffer

Our IT had the idea to change the access privileges for the DNA Master folder and all internal folders so that logged in users did not need admin privileges to make changes to anything inside those folders. Out systems are Windows 7 and this is mostly working, I would say auto-update is now working on about 95% of the computers, so there is a way to configure Windows 7 so that you don't need admin privilege. Might be worth asking your IT if this is a viable solution for your systems.

Link to this post | posted 10 Feb, 2017 15:59
cdshaffer	Yes the old tRNA-scan-SE site was shut down as of last November. You will need to follow the updated protocol which is on the seaphages faculty info page, Bioinformatics section: Click here for Revised tRNA evaluation protocol (Section 9.5).

Link to this post | posted 08 Feb, 2017 21:42

cdshaffer

Ok well it appears that flaverint does not crash my current "development" version of Starterator. So I can generate a whole phage report using it. The problem with the development version is that I have taken out much of the code that generates the summary text below the graph in anticipation of adding improved versions along the lines of the improvements to the single pham reports. Consequently the whole phage report that comes out has all the graphics but none of the summary text. So, no counting which starts are annotated and/or conserved and no list of base locations for the various starts in the phage being analyzed, this means the online version is still better for all that info. The good news is that these whole phage reports do have the part of the PDF that you need to upload to PECAAN and PECAAN will have the links to the online versions. Below are links to download the whole phage report for the crashing phage mentioned above as generated by the development version of starterator (for anyone interested the current development version is this branch of my person github repo). Whole phage pdf report for Aggie using development version of Starterator Whole phage pdf report for Flaverint using development version of Starterator Whole phage pdf report for Mattes using development version of Starterator Whole phage pdf report for Andies using development version of Starterator Whole phage pdf report for Philonius using development version of Starterator Whole phage pdf report for Aroostook using development version of Starterator Edited 08 Feb, 2017 22:21

Link to this post | posted 08 Feb, 2017 18:33

cdshaffer

OK looking into all these reports for phages aggie, philonious and Mattes, it appears these are phams that crash when you do a whole phage report but work when you ask for the individual phams. Very curious bug that will likely need time to solve. I would recommend that you use the pre-calculated web pages: for example pham 21476 results are available here: http://phages.wustl.edu/starterator/Pham21476Report.pdf Any other pham is available by just changing the pham number in the address above, so to get to pham 123 the URL would be: http://phages.wustl.edu/starterator/Pham123Report.pdf The issue I guess is PECAAN which requires a whole phage report, anyone using PECAAN let me know, I will work with you to find the best work around until I can find and fix the bug. Sorry but since this is not my own code it can be difficult for me to find bugs so I am not sure if it will be easy or difficult to solve this issue. Edited 08 Feb, 2017 21:22

Link to this post | posted 08 Feb, 2017 18:09
cdshaffer	Nick Pham 8250 does not crash when I use the SEA 2017 VM, so need a bit more background. Are you using the SEA 2017 VM or something else? Were you doing a single pham analysis of 8250 or a whole phage report on Flavrint and it crashed on gene 18?

Link to this post | posted 03 Feb, 2017 17:55

cdshaffer

If you are just creating a database that will not be updated then I would recommend just learning a bit of mysql. You might want to check out the mysql help sheet we created at the faculty retreat: https://docs.google.com/document/d/1SUUTojfDKwXCejGY1k14aY92hCqhe8XbDd_-qQPt_gQ/edit?usp=sharing and then just manually update the clusters using command like this: mysql> update phage set Cluster=”S” where Name=”Marvin”; just replace the bold parts with the cluster name and phage name in your database. If you think you will be constantly updating the phamerator database with phamdb I would focus on learning a bit of command line to run the update_clusters.py script (found in the plugins folder of phamerator) which needs a file with phage names and clusters. You can run the script each time after an update with phamdb.