BLAST - Revision history

Rf at 14:48, 9 November 2017

2017-11-09T14:48:12Z

2017-10-05T16:24:07Z

2017-10-05T16:23:56Z

2016-11-25T15:19:57Z

2016-11-24T11:09:33Z

2016-11-24T10:24:58Z

2016-05-24T14:30:53Z

2016-05-24T14:26:51Z

2016-05-15T21:05:48Z

2016-05-15T21:04:48Z

@@ Line 45: / Line 45: @@
 A typical run of the program would be:
-  makeblastdb -dbtype nucl nt
+  makeblastdb -dbtype nucl -in nt
 <ins>Explanation:</ins>
 * the -dbtype option will be prot if dealing with proteins
 * the name of the database (it's convenient if the file name is the same) must then follow.
 * By default, and if big enough size, the database is split into 1G chunks. This can be controlled most probably.

@@ Line 7: / Line 7: @@
   /shelf/public/blastntnr/preFormattedNCBI
-ALso inside this directory is the '''update_blasdb.pl''' script
+ALso inside this directory is the '''update_blastdb.pl''' script
 The preformatted version is split into 1GB chunks with index files also produced, so makeblastdb has already been run on them.

@@ Line 1: / Line 1: @@
 This is probably the best known of all bioinformatics applications, and consequently has various different aspects to it.
 =General blast issues=
@@ Line 31: / Line 41: @@
      [-max_file_sz number_of_bytes] [-taxid TaxID] [-taxid_map TaxIDMapFile]
      [-logfile File_Name] [-version]
 == blast+ ==

@@ Line 118: / Line 118: @@
 NCBI's ftp site is a good and quick place to obtain the latest up-to-date sequence databases. Logging in as anonymous, you are asked for an email address as the password.
 * ftp.ncbi.nlm.nih.gov/blast/db''', this is where you'll find the pre-formatted databases
-* ftp.ncbi.nlm.nih.gov/blast/db/FASTA''', here is where the raw unfromatted databases are
+* ftp.ncbi.nlm.nih.gov/blast/db/FASTA''', here is where the raw unformatted databases are.

@@ Line 118: / Line 118: @@
 NCBI's ftp site is a good and quick place to obtain the latest up-to-date sequence databases. Logging in as anonymous, you are asked for an email address as the password.
 * ftp.ncbi.nlm.nih.gov/blast/db''', this is where you'll find the pre-formatted databases
-* ftp.ncbi.nlm.nih.gov/blast/db/fasta''', here is where the raw unfromatted databases are
+* ftp.ncbi.nlm.nih.gov/blast/db/FASTA''', here is where the raw unfromatted databases are

← Older revision		Revision as of 10:24, 24 November 2016
Line 113:		Line 113:
	qsub mpiblastjobscript.sh		qsub mpiblastjobscript.sh
	* Note this can be easily converted to a blastp or blastx search by specifying those as programs and changing '''nt''' to '''nr'''.		* Note this can be easily converted to a blastp or blastx search by specifying those as programs and changing '''nt''' to '''nr'''.
		+
		+	= Database maintenance =
		+
		+	NCBI's ftp site is a good and quick place to obtain the latest up-to-date sequence databases. Logging in as anonymous, you are asked for an email address as the password.
		+	* ftp.ncbi.nlm.nih.gov/blast/db''', this is where you'll find the pre-formatted databases
		+	* ftp.ncbi.nlm.nih.gov/blast/db/fasta''', here is where the raw unfromatted databases are

← Older revision		Revision as of 14:30, 24 May 2016
Line 108:		Line 108:
	export MPIBLAST_LOCAL=/shelf/public/blastntnr/mpiblast46frags/nt		export MPIBLAST_LOCAL=/shelf/public/blastntnr/mpiblast46frags/nt
	mpirun -np $NSLOTS $MPIBLASTBIN -d nt -i $1 -p blastn -e 0.000001 -K 1 -m 9 -o $2		mpirun -np $NSLOTS $MPIBLASTBIN -d nt -i $1 -p blastn -e 0.000001 -K 1 -m 9 -o $2
		+
		+	<ins>Explanation</ins>:
		+	This script takes two arguments, so, if its name is '''mpiblastjobscript.sh''' it will be run with
		+	qsub mpiblastjobscript.sh
		+	* Note this can be easily converted to a blastp or blastx search by specifying those as programs and changing '''nt''' to '''nr'''.

@@ Line 74: / Line 74: @@
 Therefore the database need to be fragmented and also - as is usual in blast - formatted.
-The number of fragments is two less than the number of processes to be used. So, for 64 processes, the database will need to be fragmented into 62 parts. A script for doing this is as follows:
+The number of fragments is two less than the number of processes to be used. So, for 48 processes, the database will need to be fragmented into 46 parts. A script for doing this is as follows:
   #!/bin/bash
@@ Line 92: / Line 92: @@
 As you can see, this uses up a good deal of temporary hard disk space. There is an alternative way by using "zcat" and a pipe, but this also names the database fragments as "stdin" which is quite inconvenient.

@@ Line 66: / Line 66: @@
 Note here the '''-max_target_seqs''' value of just 1, which will return only the best hit. The output format '''-outfmt''' of 7 is the text tabular for mat which is the easiest to view,
-and has added comments which can easily be deleted, but are informative.
+and has added comments which can easily be deleted, but are informative. This line is also valid for blastp.
 =mpiBLAST=

@@ Line 60: / Line 60: @@
      [-num_alignments int_value] [-html] [-max_target_seqs num_sequences]
      [-num_threads int_value] [-remote] [-version]
 =mpiBLAST=