cmd

Documentation

Index

• Constants
• Variables
• func CalcSignatureSize(numElements uint64, numHashes int, falsePositiveRate float64) uint64
• func Execute()
• func MergeUnikIndex(opt *Options, prefix string, files []string, outFile string) error
• func ParseByteSize(val string) (int, error)
• type Options
• type UnikFileInfo
  • func (i UnikFileInfo) String() string
• type UnikFileInfos
  • func (l UnikFileInfos) Len() int
  • func (l UnikFileInfos) Less(i int, j int) bool
  • func (l UnikFileInfos) Swap(i int, j int)
• type UnikIndex
  • func NewUnixIndex(file string, useMmap bool) (*UnikIndex, error)
  • func (idx *UnikIndex) Close() error
  • func (idx *UnikIndex) Search(hashes [][]uint64, queryCov float64, targetCov float64) map[string][3]float64
  • func (idx *UnikIndex) String() string
• type UnikIndexDB
  • func NewUnikIndexDB(path string, useMmap bool) (*UnikIndexDB, error)
  • func (db *UnikIndexDB) Close() error
  • func (db *UnikIndexDB) Search(kmers []uint64, threads int, queryCov float64, targetCov float64) map[string][3]float64
  • func (db *UnikIndexDB) SearchMap(kmers map[uint64]interface{}, threads int, queryCov float64, targetCov float64) map[string][3]float64
  • func (db *UnikIndexDB) String() string
• type UnikIndexDBInfo
  • func NewUnikIndexDBInfo(files []string) UnikIndexDBInfo
  • func UnikIndexDBInfoFromFile(file string) (UnikIndexDBInfo, error)
  • func (i UnikIndexDBInfo) Check() error
  • func (i UnikIndexDBInfo) String() string
  • func (i UnikIndexDBInfo) WriteTo(file string) error

Constants

const PosPopCountBufSize = 128

PosPopCountBufSize defines the buffer size of byte slice feeding to pospopcount (github.com/clausecker/pospop).

Theoretically, size >240 is better, but in this scenario, we need firstly transposing the signature matrix, which is the performance bottleneck. Column size of the matrix is fixed, therefore we must control the row size to balance time of matrix transposing and popopcount.

128 is the best value for my machine (AMD ryzen 2700X).

const UnikIndexDBVersion uint8 = 2

Variables

var BufferSize = 65536 //os.Getpagesize()

BufferSize is size of buffer

var ErrVersionMismatch = errors.New("unikmer/index-db: version mismatch")

var RootCmd = &cobra.Command{
	Use:   "unikmer",
	Short: "Unique-Kmer Toolkit",
	Long: fmt.Sprintf(`unikmer - Unique-Kmer Toolkit

unikmer is a toolkit for nucleic acid k-mer analysis, providing functions
including set operation, indexing, and searching on k-mers optional with
TaxIDs but without count information.

K-mers are either encoded (k<=32) or hashed (arbitrary k) into 'uint64',
and serialized in binary file with extension '.unik'.

TaxIDs can be assigned when counting k-mers from genome sequences,
and LCA (Lowest Common Ancestor) is computed during set opertions
including computing union, intersecton, set difference, unique and
repeated k-mers.

Version: v%s

Author: Wei Shen <shenwei356@gmail.com>

Documents  : https://shenwei356.github.io/unikmer
Source code: https://github.com/shenwei356/unikmer

Dataset (optional):

  Manipulating k-mers with TaxIDs needs taxonomy file from e.g., 
  NCBI Taxonomy database, please extract "nodes.dmp", "names.dmp",
  "delnodes.dmp" and "merged.dmp" from link below into ~/.unikmer/ ,
  ftp://ftp.ncbi.nih.gov/pub/taxonomy/taxdump.tar.gz , 
  or some other directory, and later you can refer to using flag
  --data-dir or environment variable UNIKMER_DB.

  For GTDB, use https://github.com/nick-youngblut/gtdb_to_taxdump 
  for taxonomy convertion.

  Note that TaxIDs are represented using uint32 and stored in 4 or
  less bytes, all TaxIDs should be in range of [1, %d]

`, VERSION, maxUint32),
}

RootCmd represents the base command when called without any subcommands

var VERSION = "0.12.1"

VERSION is the version

Functions

func CalcSignatureSize

func CalcSignatureSize(numElements uint64, numHashes int, falsePositiveRate float64) uint64

From https://github.com/bingmann/cobs/blob/master/cobs/util/calc_signature_size.cpp

func Execute

func Execute()

Execute adds all child commands to the root command sets flags appropriately. This is called by main.main(). It only needs to happen once to the rootCmd.

func MergeUnikIndex

func MergeUnikIndex(opt *Options, prefix string, files []string, outFile string) error

func ParseByteSize

func ParseByteSize(val string) (int, error)

ParseByteSize parses byte size from string.

Types

type Options

type Options struct {
	NumCPUs          int
	Verbose          bool
	Compress         bool
	Compact          bool
	CompressionLevel int
	MaxTaxid         uint32
	IgnoreTaxid      bool
	DataDir          string
	NodesFile        string
	CacheLCA         bool

	NoCheckFile bool
}

Options contains the global flags

type UnikFileInfo

type UnikFileInfo struct {
	Path  string
	Name  string
	Kmers int64
}

func (UnikFileInfo) String

func (i UnikFileInfo) String() string

type UnikFileInfos

type UnikFileInfos []UnikFileInfo

func (UnikFileInfos) Len

func (l UnikFileInfos) Len() int

func (UnikFileInfos) Less

func (l UnikFileInfos) Less(i int, j int) bool

func (UnikFileInfos) Swap

func (l UnikFileInfos) Swap(i int, j int)

type UnikIndex

type UnikIndex struct {
	Path   string
	Header index.Header
	// contains filtered or unexported fields
}

func NewUnixIndex

func NewUnixIndex(file string, useMmap bool) (*UnikIndex, error)

func (*UnikIndex) Close

func (idx *UnikIndex) Close() error

func (*UnikIndex) Search

func (idx *UnikIndex) Search(hashes [][]uint64, queryCov float64, targetCov float64) map[string][3]float64

func (*UnikIndex) String

func (idx *UnikIndex) String() string

type UnikIndexDB

type UnikIndexDB struct {
	Info   UnikIndexDBInfo
	Header index.Header

	Indices []*UnikIndex
	// contains filtered or unexported fields
}

func NewUnikIndexDB

func NewUnikIndexDB(path string, useMmap bool) (*UnikIndexDB, error)

func (*UnikIndexDB) Close

func (db *UnikIndexDB) Close() error

func (*UnikIndexDB) Search

func (db *UnikIndexDB) Search(kmers []uint64, threads int, queryCov float64, targetCov float64) map[string][3]float64

func (*UnikIndexDB) SearchMap

func (db *UnikIndexDB) SearchMap(kmers map[uint64]interface{}, threads int, queryCov float64, targetCov float64) map[string][3]float64

func (*UnikIndexDB) String

func (db *UnikIndexDB) String() string

type UnikIndexDBInfo

type UnikIndexDBInfo struct {
	Version      uint8    `yaml:"version"`
	IndexVersion uint8    `yaml:"unikiVersion"`
	K            int      `yaml:"k"`
	Hashed       bool     `yaml:"hashed"`
	Canonical    bool     `yaml:"canonical"`
	NumHashes    int      `yaml:"hashes"`
	FPR          float64  `yaml:"fpr"`
	BlockSize    int      `yaml:"blocksize"`
	Kmers        int      `yaml:"totalKmers"`
	Files        []string `yaml:"files"`
	NumNames     int      `yaml:"numNames"`
	Names        []string `yaml:"names"`
	Sizes        []uint64 `yaml:"kmers"`
	// contains filtered or unexported fields
}

func NewUnikIndexDBInfo

func NewUnikIndexDBInfo(files []string) UnikIndexDBInfo

func UnikIndexDBInfoFromFile

func UnikIndexDBInfoFromFile(file string) (UnikIndexDBInfo, error)

func (UnikIndexDBInfo) Check

func (i UnikIndexDBInfo) Check() error

func (UnikIndexDBInfo) String

func (i UnikIndexDBInfo) String() string

func (UnikIndexDBInfo) WriteTo

func (i UnikIndexDBInfo) WriteTo(file string) error