Hi,
I am carrying out functional annotation of predicted genes in our genome using AHRD.
I have the same problem as stated here.
The example runs correctly. I have used the example .yml file and replaced example files with our files, but it fails to run with the following error:
source jre-7.11;java -jar ./dist/ahrd.jar batch1.yml
Usage: java -Xmx2g -jar ahrd.jar input.yml
Started AHRD... ...initialised proteins in 0sec, currently occupying 10 MB We are sorry, an un-expected ERROR occurred: java.lang.NullPointerException at
org.biojava.bio.program.sax.BlastSAXParser.interpret(BlastSAXParser.java:224) at
org.biojava.bio.program.sax.BlastSAXParser.parse(BlastSAXParser.java:168) at
org.biojava.bio.program.sax.BlastLikeSAXParser.onNewDataSet(BlastLikeSAXParser.java:314) at
org.biojava.bio.program.sax.BlastLikeSAXParser.interpret(BlastLikeSAXParser.java:276) at
org.biojava.bio.program.sax.BlastLikeSAXParser.parse(BlastLikeSAXParser.java:163) at
ahrd.model.BlastResult.parseBlastResults(Unknown Source) at
ahrd.controller.AHRD.parseBlastResults(Unknown Source) at
ahrd.controller.AHRD.setup(Unknown Source) at
ahrd.controller.AHRD.main(Unknown Source)
The only thing different is that the example pairwise input files have been produced using an earlier version of BLAST, but I can't see this being the problem as it's been used in a number of recent papers (e.g. http://www.plosgenetics.org/article/info%3Adoi%2F10.1371%2Fjournal.pgen.1004078#pgen-1004078-g005 & the wheat genome), and I doubt they used an earlier version of BLAST.
Despite this, the error message does allude to some problem with the BLAST parsing - it seems AHRD was presented in 2013 (https://pag.confex.com/pag/xxi/webprogram/Paper5912.html), yet the suggested blast search in the AHRD manual is:
blastall -p blastp -i proteins.fasta -o swissprot_blastout.pairwise -d swissprot.fasta -e 0.0001 -v 200 -b 200 -m 0 ...
i.e. not BLAST+.
Does anyone have a solution/managed to get this program working (without using the old version of BLAST - if this is indeed the problem)?
Thanks
I ran blastp from the blast+ suite using the example protein.fasta file (against swissprot, trembl, tair).
Swapping in these files in place of the blastall pairwise output files (provided as example data) produced the error seen above.
So I think AHRD requires pairwise files generated from blastall.