FindSynteny {DECIPHER} | R Documentation |
Finds syntenic blocks between groups of sequences in a database.
FindSynteny(dbFile, tblName = "Seqs", identifier = "", useFrames = TRUE, alphabet = AA_REDUCED[[1]], geneticCode = GENETIC_CODE, sepCost = 0, gapCost = -0.01, shiftCost = 0, codingCost = 0, maxSep = 2000, maxGap = 5000, minScore = 30, dropScore = -100, maskRepeats = TRUE, allowOverlap = TRUE, storage = 0.5, processors = 1, verbose = TRUE)
dbFile |
A SQLite connection object or a character string specifying the path to the database file. |
tblName |
Character string specifying the table where the sequences are located. |
identifier |
Optional character string used to narrow the search results to those matching a specific identifier. If |
useFrames |
Logical specifying whether to use 6-frame amino acid translations to help find more distant hits. Using the |
alphabet |
Character vector of amino acid groupings used to reduce the 20 standard amino acids into smaller groups. Alphabet reduction helps to find more distant homologies between sequences. A non-reduced amino acid alphabet can be used by setting |
geneticCode |
Either a character vector giving the genetic code to use in translation, or a list containing one genetic code for each identifier. If a list is provided then it must be named by the corresponding identifiers in the database. |
sepCost |
Cost per nucleotide separation between hits to apply when chaining hits into blocks. |
gapCost |
Cost for gaps between hits to apply when chaining hits into blocks. |
shiftCost |
Cost for shifting between different reading frames when chaining reduced amino acid hits into blocks. |
codingCost |
Cost for switching between coding and non-coding hits when chaining hits into blocks. |
maxSep |
Maximal separation (in nucleotides) between hits in the same block. |
maxGap |
The maximum number of gaps between hits in the same block. |
minScore |
The minimum score required for a chain of hits to become a block. Higher values of |
dropScore |
The change from maximal score required to stop extending blocks. |
maskRepeats |
Logical specifying whether to “soft” mask repeats when searching for hits. |
allowOverlap |
Logical specifying whether to permit blocks to overlap on the same sequence. |
storage |
Excess gigabytes available to store objects so that they do not need to be recomputed in later steps. This should be a number between zero and a (modest) fraction of the available system memory. Note that more than |
processors |
The number of processors to use, or |
verbose |
Logical indicating whether to display progress. |
Long nucleotide sequences, such as genomes, are often not collinear or may be composed of many smaller segments (e.g., contigs). FindSynteny
searches for “hits” between sequences that can be chained into collinear “blocks” of synteny. Hits are defined as k-mer exact nucleotide matches or k-mer matches in a reduced amino acid alphabet (if useFrames
is TRUE
). Hits are chained into blocks as long as they are: (1) within the same sequence, (2) within maxSep
and maxGap
distance, and (3) help maintain the score above minScore
. Blocks are extended from their first and last hit until their score drops below dropScore
from the maximum that was reached. This process results in a set of hits and blocks stored in an object of class “Synteny”.
An object of class “Synteny”.
FindSynteny
is intended to be used on sets of sequences with up to ~100 million nucleotides total per identifier. For this reason, better performance can sometimes be achieved by assigning a unique identifier to each chromosome belonging to a large genome.
Erik Wright eswright@pitt.edu
db <- system.file("extdata", "Influenza.sqlite", package="DECIPHER") synteny <- FindSynteny(db) synteny pairs(synteny) # scatterplot matrix