homologysearch.txt

比对算法的具体应用DNA序列分析 ——基因序列 ——基因表达调控信息 寻找基因牵涉到两个方面的工作 ： 识别与基因相关的特殊序列信号 预测基因的编码区域 结合两个方面的结果确定基因的位置和结构 基因表

***  GENOME EXPLORER HELP FILE  ***

To provide help on using the GUI, and information about how the programs run

Contents

1)  Outline of Function
2)  Parameters loaded from the .inf file (settings menu)
3)  The User Interface
4)  Underlying Method


---  HOMOLOGY SEARCH  ---

---  1)  Outline of Function ---

To search a phylogeny for homologous sequences.

Uses blast to search for homologues - using the evalue and score criteria
set by the user.

Produces an output file describing the input data, and listing
the homologues.  Also produces a namelist file to enable the user to refer back
from the numerical naming system in the homology file to the 
sequence descriptions from the input fasta files

REQUIRES chromosome fasta files to have sequences in the CORRECT ORDER



---  2)  Parameters loaded from the .inf file (settings menu) ---

blastExePath          path to the blastall executable (blastall.exe)

formatdbExePath       path to the formatdb executable (formatdb.exe)

fastacmdExePath       path to the fastacmd executable (fastacmd.exe)

blastOutfileExt       file extension for the blast report (.blst)

blastParserFilename   default outfile name (not path) for the summary file of parsed blast
                      reports for an entire fasta file of sequences (blastParsed.sum)
                      Will be written to blastParserOutDir
                      
blastTempSeqFilename  filename (not path) for the single sequence file to be overwritten each
                      time a sequence from a fasta file gets blasted.  Will be written to 
                      blastTemSeqDir ("seq.temp")
                      
blastTempSeqDir       directory in which to write single sequences pulled from fasta files so
                      that they can be blasted individually
                      
blastTempDbFilename   filename (not path) for a temporary database (if one needs to be
                      created from a fasta file).  Will refer to several files with different
                      extensions, so does NOT need an extension here (dbtemp)
                      
blastTempDbDir        directory in which to write a temporary blastable database (if one
                      needs to be created from a fasta file)

homolExt              file extension for files produced by this program (.homol)

homolOutDirPath       default directory in which to write the final outfile
                      (text representing a Homology object - infile for geneOrderMatrix program)

outDirPath            default directory in which to write all files produced in generating the
                      final homology file

genomeDirPath         directory in which to open file chooser when browsing for phylogenies 
                      / chromosome.fasta files 

---  3)  The User Interface  ---

phylogenies generated by evolve / 
user defined phylogenies (evolve directory structure) / 
user defined phylogenies 
    radio buttons to switch between layouts for data entry in the "select phylogenies"
    tab.  'phylogenies generated by evolve' provides a simple entry form
    because the data is output in a consistent format.  This format is a directory for 
    each phylogeny, within that, a directory for each species, within that, a fasta file
    for each chromosome - with sequences listed in the order in which they appear on the
    chromosome.  The user defined phylogenies (evolve directory structure) allows the user
    to take advantage of this directory structure to easily enter their own data.

BASIC PARAMETERS TAB

protein sequences / dna sequences
    radio button switch indicating the type of sequences being used.  This effects the
    blast program used to look for homologues (blastp or blastn)
    
cut-off evalue 
    the maximum evalue of a hit that is considered a homologue 
    (must also satisfy score criteria)

cut-off score
    the minimum score of a hit that is considered a homologue
    (must also satisfy evalue criteria)

only include genes with homologues in output
    If the homology file is going to be used as a basis for a phylogeny, it simplifies
    things if only the genes with at least one homologue (somewhere amongst all 
    other chromosomes) be included in the final file.

evolve input - use only genes
    Evolve has several different types of sequence - genes, introns, transposons and 
    footprints.  ticking this box allows the program to ignore anything that isn't a 
    gene when building the homology file.

evolve input - parse by seq names (no blast)
    If the evolve files have full evolve names that enable the ancestry of each gene to
    be calculated from the name, this option uses that information and will run a lot
    quicker than the blast method.

select directory to output final homology file
    select the directory to which the final homology file will be written

select directory for all output files
    An inner directory will be created within this one, to hold only files output by
    this run of the program.  This includes all blast reports, parsed blast reports, 
    AND the TRANSLATION files to enable gene ids in the final homology file to be
    related back to actual sequence descriptions.

enter name for final homology file
    the name (not path) of the final homology file.  An extension (set below)
    will be added.  If you are writing all homology files to the same directory over
    several runs of this program, ensure you name this file so as not to overwrite
    any others.

enter extension for final homology file
    the extension that will be added to the final homology file (defaults to 
    homolExt in settings)
    

SELECT PHYLOGENIES (ANY) TAB

add / delete species  
    enter / remove species name in species list box.  Edit name to something
    meaningful!

add chromosome
    add path to a fasta file containing all sequences in a chromosome, in the order
    in which they occur.  The ORDERING is VERY IMPORTANT.
    Edit the name of the chromosome to something meaningful!
    Will not allow two chromosomes with identical paths (i.e. the same file) to be
    added twice - even in different species.

delete selected chromosome
    removes the selected chromosome from the list

SELECT PHYLOGENIES (EVOLVE STRUCTURE) TAB

select / remove directories
    these are directories with evolved genomes in.  If unaltered evolve output, the
    directories will have names like phy_1, phy_2 etc.
    Each phylogeny is analysed independently, and a homology file will be produced 
    for each in turn.  In this case final homology files will be named with the name
    of the phylogeny directory with the homology extension added (e.g. phy_2.homol).
 

---  4)  Underlying Method   ---


 Class to search a phylogeny for homologous sequences.
 
 The search is conducted using RunBlast and RunBlastParser objects.  A homologous
 sequences are determined using cut off evalues and scores.
 
 Every chromosome to be included in the phylogeny must be in a separate file, 
 and all sequences must appear in the same order in the fasta file as they do on 
 the chromosome.
 
 A single fasta file (allSeq.fasta) is constructed from all the chromosomes in all 
 the species in the phylogeny.  Each sequence is renamed according to its species, 
 chromosome and chromosome position in a way that the program can easily decipher.  
 A translation file is written to enable the user to relate a program generated sequence 
 id back to the original sequence name (seqNameList.txt).
 
 Every sequence in allSeq.fasta is then blasted against a database consisting 
 only of the sequences in allSeq.fasta.  The top hit of every sequence will be to 
 itself in the database.
 
 Each blast report is parsed to produce a new file listing only the program 
 assigned name of significant hits, the hit start position and the hit stop position.  
 A hit is judged to be significant if it meets the e-value and score criteria specified 
 in the HomologySearchParameters object.
 
 The parsed blast reports are used to construct a homology file.  Each parsed report 
 in turn is read by the program. All sequences in a single parsed file are assigned 
 the same homologueId, and a direction value based on the hit start position and hit 
 stop position (if start position is less than stop position the hit sequence ran in the 
 same direction as the query sequence and is assigned the value 'true' for its direction.  
 Otherwise it is assigned the value 'false').
 
 If a sequence has appeared before in a previous parsed file, all sequences in the 
 current file are assigned the same homologueId as that sequence.  All direction 
 values are adjusted accordingly (if sequence A already appears in the homologue 
 list with a homologueId of 5 and a direction of false, but here has a 
 provisional homologueId of 12 and a direction of true, all sequences in this 
 file are reassigned the homologueId 5 and their directions are reversed).  This 
 ensures the homologue list is always internally consistent.
 
 The final output file is a list of all sequences in the phylogeny detailing 
 speciesNumber, chromosomeNumber, positionOnChromosome, homologueId and direction. 
 A translation section is provided to record the species name corresponding to each 
 speciesNumber and the chromosome name corresponding to each chromosomeNumber.  
 To identify an individual sequence by name, the user must refer to the seqNameList.txt 
 file produced when the sequences are originally renamed.
 
 This class creates a new directory based on date and time in which to write all temporary and
 in-process files.