Differences

This shows you the differences between two versions of the page.

--- blast_and_plast [2021/09/02 15:59] – 38.20.199.40
+++ blast_and_plast [2024/11/04 10:14] (current) – 110.239.172.216
@@ Line 10: / Line 10: @@
 cd $PWD
 CPUs=10
-DB=/db1/nr-nt-fasta-oct-2020/nt
+DB=/scratch3/rogerlab_databases/other_dbs/nr_March252023/plast/nr.fasta
 QF=yourquery.fasta
-plast -e 1e-10 -max-hit-per-query 1 -outfmt 1 -a $CPUs -p plastn -max-database-size 10000000000 -i $QF -d $DB -o $QF.plout -force-query-order 1000
+plast -e 1e-10 -max-hit-per-query 1 -outfmt 1 -a $CPUs -p plastp -max-database-size 10000000000 -i $QF -d $DB -o $QF.plout -force-query-order 1000
 </code>
-to parse the output see http://129.173.88.134:81/dokuwiki/doku.php?id=dayana_salas_-_utility_scripts_taxonomy_coloring_trees_phylogenetics_mixture_models_domain_architecture_and_more
+to parse the output see https://perun.biochem.dal.ca/user-wiki/doku.php?id=taxonomy_recovery
@@ Line 26: / Line 26: @@
 #$ -S /bin/bash
 . /etc/profile
-#$ -pe threaded 1
+#$ -pe threaded 10
 #$ -cwd
 source activate blast
-export BLASTDB=/db1/nr-nt-oct-2020-v5/
+export BLASTDB=/db1/blast-may-2024/
 DB=nt
 query=your_query.fasta
-blastn -db $DB -query $query -out yourqueryresults.blout -num_threads 1 -outfmt "6 qseqid sseqid evalue pident qcovs length slen qlen qstart qend sstart send stitle"
+blastn -db $DB -query $query -out yourqueryresults.blout -num_threads 10 -outfmt "6 qseqid sseqid evalue pident qcovs length slen qlen qstart qend sstart send stitle"
-source deactivate
+conda deactivate
 </code>
-Both shells using NCBI nt database (/db1/nr-nt-jan-2019/nt.nal), but the formats for specifying DB are different for BLAST and PLAST.
+Both shells use NCBI nt database, but PLAST doesn't support new v5 NCBI nr and nt databases and can cause Segmentation fault error.
-**Guide for BLAST usage**
+<Last updated by Dandan Zhao on Jun 11, 2024>
-  - blastp:search protein database(e.g., SwissProt db, NCBI-nr) using protein sequence query
-  - blastn:search nucleotide database(e.g., NCBI-nt, MMETSP_DB_clean.v2018.fa)using nucleotide sequence query
-  - blastx:search protein database with translated nucleotide sequence query
-  - tblastn:search translated nucleotide database with protein sequence query
-  - tblastx:search translated nucleotide database with translated nucleotide sequence query
-//Note: blastp and blastx can usually provide better hit alignments than blastn, especially for distantly related species.This is because amino acids sequences are more conserved than nucleotides (Koonin and Galperin, 2002).//
-**General bugs**
-when mistakenly use blast options(e.g., blastn or blastp) or query sequence (amino acids or nucleotides sequences):
-<code>
-Error 1:
-FASTA-Reader: Ignoring invalid residues at position(s): On line 7: 4, 8, 10, 13, 27-29, 32, 42, 45, 51, 53, 56, 63, 66-67, 70, 78
-FASTA-Reader: Ignoring invalid residues at position(s): On line 8: 6, 9, 15, 19-20, 22, 28, 34-39, 45-48, 52
-</code>
-Solve :
-This is due to mistakenly using the blast options.
-<code>
-Error 2:
-BLAST Database error: No alias or index file found for protein database [XXX.fa] in search path [/misc/scratch2/XXX:]
-</code>
-Solve 2:
-This is due to mistakenly treating nucleotide database as protein database.
-**Parsing Blast results**
-Using BLASTP search option to blast the amino acid sequences against uniport_db database.
-<code>
-> ./blastp -query XXX.fasta -db uniprot_db -out BLASTP_XXX_uniprot.xml -evalue 1e-5 -outfmt 5
-</code>
-The **BLAST XML file** (-outfmt 5) can include useful information comparing to the BLAST Tabular file (-outfmt 6), such as the aligned sequence, the sequence of the hit, and the description of hits into the database. However, the XML format is not human-readable.
-Users will need to employ a commonly used parser (//Blastxml_to_tabular.py//) from the link(https://github.com/peterjc/galaxy_blast/blob/master/tools/ncbi_blast_plus/blastxml_to_tabular.py)(Cock et al.,2015), which is a custom python script, to convert a BLAST XML file to a desired tabular output (tab-delimited file).
-<code>
-python blastxml_to_tabular.py -c qseqid,qlen,salltitles,sseqid,slen,bitscore,qframe,pident,evalue,qstart,qend,sstart,send,length BLASTP_XXX_uniprot.xml > BLASTP_XXX_uniprot.tsv
-</code>
-"-c" option is to infer the displayed columns, there are 25 columns in total. User can infer the front 12 columns via "-c std"; infer 25 columns via "-c ext";infer the personalized via using selected column name delimited by comma "-c qseqid,sseqid"
-<code>
-qseqid    Query Seq-id (ID of your sequence)
-sseqid    Subject Seq-id (ID of the database hit)
-pident    Percentage of identical matches
-length    Alignment length
-mismatch  Number of mismatches
-gapopen   Number of gap openings
-qstart    Start of alignment in query
-qend      End of alignment in query
-sstart    Start of alignment in subject (database hit)
-send      End of alignment in subject (database hit)
-evalue    Expectation value (E-value)
-bitscore  Bit score
-sallseqid     All subject Seq-id(s), separated by ';'
-score         Raw score
-nident        Number of identical matches
-positive      Number of positive-scoring matches
-gaps          Total number of gaps
-ppos          Percentage of positive-scoring matches
-qframe        Query frame
-sframe        Subject frame
-qseq          Aligned part of query sequence
-sseq          Aligned part of subject sequence
-qlen          Query sequence length
-slen          Subject sequence length
-salltitles    All subject titles, separated by '&lt;&gt;'
-</code>
-<code>
-    $ python blastxml_to_tabular.py -o output.tabular -c std input.xml
-    $ python blastxml_to_tabular.py -o output.tabular -c ext input.xml
-    $ python blastxml_to_tabular.py -o output.tabular -c qseqid,qlen,salltitles,sseqid,slen,bitscore,qframe,pident,evalue,qstart,qend,sstart,send,length input.xml
-</code>