#LyX 2.0 created this file. For more info see http://www.lyx.org/
\lyxformat 413
\begin_document
\begin_header
\textclass report
\begin_preamble
\usepackage[letterpaper]{geometry}
\usepackage{siunitx}
\end_preamble
\use_default_options true
\begin_modules
theorems-ams
eqs-within-sections
figs-within-sections
\end_modules
\maintain_unincluded_children false
\language english
\language_package default
\inputencoding auto
\fontencoding global
\font_roman default
\font_sans default
\font_typewriter default
\font_default_family default
\use_non_tex_fonts false
\font_sc false
\font_osf false
\font_sf_scale 100
\font_tt_scale 100

\graphics default
\default_output_format default
\output_sync 0
\bibtex_command default
\index_command default
\paperfontsize default
\spacing single
\use_hyperref false
\papersize default
\use_geometry false
\use_amsmath 1
\use_esint 1
\use_mhchem 1
\use_mathdots 1
\cite_engine basic
\use_bibtopic false
\use_indices false
\paperorientation portrait
\suppress_date false
\use_refstyle 1
\index Index
\shortcut idx
\color #008000
\end_index
\secnumdepth 2
\tocdepth 2
\paragraph_separation indent
\paragraph_indentation default
\quotes_language english
\papercolumns 1
\papersides 1
\paperpagestyle default
\tracking_changes false
\output_changes false
\html_math_output 0
\html_css_as_file 0
\html_be_strict false
\end_header

\begin_body

\begin_layout Title
Jellyfish 2 User Guide
\end_layout

\begin_layout Standard
\begin_inset CommandInset toc
LatexCommand tableofcontents

\end_inset


\end_layout

\begin_layout Standard
\begin_inset FormulaMacro
\newcommand{\switch}[1]{\texttt{-\@-#1}}
\end_inset


\begin_inset FormulaMacro
\newcommand{\opt}[1]{\texttt{-#1}}
\end_inset


\end_layout

\begin_layout Chapter
Getting started
\end_layout

\begin_layout Section
Counting all 
\begin_inset Formula $k$
\end_inset

-mers
\end_layout

\begin_layout Standard
The basic command to count all 
\begin_inset Formula $k$
\end_inset

-mers is as follows:
\end_layout

\begin_layout LyX-Code
jellyfish count -m 21 -s 100M -t 10 -C reads.fasta
\end_layout

\begin_layout Standard
This will count canonical (
\begin_inset Formula $\opt C$
\end_inset

) 
\begin_inset Formula $21$
\end_inset

-mers (
\family sans

\begin_inset Formula $\opt m\,21$
\end_inset


\family default
), using a hash with 100 million elements (
\family sans

\begin_inset Formula $\opt s\,\unit[100]{M}$
\end_inset


\family default
) and 10 threads (
\family sans

\begin_inset Formula $\opt t\,10$
\end_inset


\family default
) in the sequences in the file 
\family sans
reads.fasta
\family default
.
 The output is written in the file '
\family sans
mer_counts.jf
\family default
' by default (change with 
\family sans

\begin_inset Formula $\opt o$
\end_inset


\family default
 switch).
 
\end_layout

\begin_layout Standard
To compute the histogram of the 
\begin_inset Formula $k$
\end_inset

-mer occurrences, use the 
\family sans
histo
\family default
 subcommand (see section
\begin_inset space ~
\end_inset


\begin_inset CommandInset ref
LatexCommand ref
reference "sec:histo"

\end_inset

):
\end_layout

\begin_layout LyX-Code
jellyfish histo mer_counts.jf
\end_layout

\begin_layout Standard
To query the counts of a particular 
\begin_inset Formula $k$
\end_inset

-mer, use the 
\family sans
query
\family default
 subcommand (see section
\begin_inset space ~
\end_inset


\begin_inset CommandInset ref
LatexCommand ref
reference "sec:query"

\end_inset

):
\end_layout

\begin_layout LyX-Code
jellyfish query mer_counts.jf AACGTTG
\end_layout

\begin_layout Standard
To output all the counts for all the 
\begin_inset Formula $k$
\end_inset

-mers in the file, use the 
\family sans
dump
\family default
 subcommand (see section
\begin_inset space ~
\end_inset


\begin_inset CommandInset ref
LatexCommand ref
reference "sec:dump"

\end_inset

):
\end_layout

\begin_layout LyX-Code
jellyfish dump mer_counts.jf > mer_counts_dumps.fa
\end_layout

\begin_layout Standard
To get some information on how, when and where this jellyfish file was generated
, use the 
\family sans
info
\family default
 subcommand (see section
\begin_inset space ~
\end_inset


\begin_inset CommandInset ref
LatexCommand ref
reference "sec:info"

\end_inset

):
\end_layout

\begin_layout LyX-Code
jellyfish info mer_counts.jf
\end_layout

\begin_layout Standard
For more detail information, see the relevant sections in this document.
 All commands understand 
\begin_inset Formula $\switch{help}$
\end_inset

 and will produce some information about the switches available.
\end_layout

\begin_layout Subsection
Counting 
\begin_inset Formula $k$
\end_inset

-mers in sequencing reads
\end_layout

\begin_layout Standard
In sequencing reads, it is unknown which strands of the DNA is sequenced.
 As a consequence, a 
\begin_inset Formula $k$
\end_inset

-mer or its reverse complement are essentially equivalent.
 The canonical representative of a 
\begin_inset Formula $k$
\end_inset

-mer 
\begin_inset Formula $m$
\end_inset

 is by definition 
\begin_inset Formula $m$
\end_inset

 or the reverse complement of 
\begin_inset Formula $m$
\end_inset

, whichever comes first lexicographically.
 The 
\family sans

\begin_inset Formula $\opt C$
\end_inset


\family default
 switch instructs to save in the hash only canonical 
\begin_inset Formula $k$
\end_inset

-mers, while the count is the number of occurrences of both a 
\begin_inset Formula $k$
\end_inset

-mer and it reverse complement.
\end_layout

\begin_layout Standard
The size parameter (given with 
\begin_inset Formula $\opt s$
\end_inset

) is an indication of the number 
\begin_inset Formula $k$
\end_inset

-mers that will be stored in the hash.
 For sequencing reads, one this size should be the size of the genome plus
 the 
\begin_inset Formula $k$
\end_inset

-mers generated by sequencing errors.
 For example, if the error rate is 
\begin_inset Formula $e$
\end_inset

 (e.g.
\begin_inset ERT
status open

\begin_layout Plain Layout


\backslash
@
\end_layout

\end_inset

 Illumina reads, usually 
\begin_inset Formula $e\approx1\%$
\end_inset

), with an estimated genome size of 
\begin_inset Formula $G$
\end_inset

 and a coverage of 
\begin_inset Formula $c$
\end_inset

, the number of expected 
\begin_inset Formula $k$
\end_inset

-mers is 
\begin_inset Formula $G+Gcek$
\end_inset

.
 This assume
\end_layout

\begin_layout Quote
NOTE: unlike in Jellyfish 1, this 
\family sans
-s
\family default
 parameter is only an estimation.
 If the size given is too small to fit all the 
\begin_inset Formula $k$
\end_inset

-mers, the hash size will be increased automatically or partial results
 will be written to disk and finally merged automatically.
 Running '
\family sans
jellyfish merge
\family default
' should never be necessary, as now jellyfish now takes care of this task
 on its own.
\end_layout

\begin_layout Standard
If the low frequency 
\begin_inset Formula $k$
\end_inset

-mers (
\begin_inset Formula $k$
\end_inset

-mers occurring only once), which are mostly due to sequencing errors, are
 not of interest, one might consider counting only high-frequency 
\begin_inset Formula $k$
\end_inset

-mers (see section
\begin_inset space ~
\end_inset


\begin_inset CommandInset ref
LatexCommand ref
reference "sec:Counting-high-frequency--mers"

\end_inset

), which uses less memory and is potentially faster.
\end_layout

\begin_layout Subsection
Counting 
\begin_inset Formula $k$
\end_inset

-mers in a genome
\end_layout

\begin_layout Standard
In an actual genome or finished sequence, a 
\begin_inset Formula $k$
\end_inset

-mer and its reverse complement are not equivalent, hence using the 
\family sans

\begin_inset Formula $\opt C$
\end_inset


\family default
 switch does not make sense.
 In addition, the size for the hash can be set directly to the size of the
 genome.
\end_layout

\begin_layout Section
Counting high-frequency 
\begin_inset Formula $k$
\end_inset

-mers
\begin_inset CommandInset label
LatexCommand label
name "sec:Counting-high-frequency--mers"

\end_inset


\end_layout

\begin_layout Standard
Jellyfish offers two way to count only high-frequency 
\begin_inset Formula $k$
\end_inset

-mers (meaning only 
\begin_inset Formula $k$
\end_inset

-mers with count 
\begin_inset Formula $>1$
\end_inset

), which reduces significantly the memory usage.
 Both methods are based on using Bloom filters.
 The first method is a one pass method, which provides approximate count
 for some percentage of the 
\begin_inset Formula $k$
\end_inset

-mers.
 The second method is a two pass method which provides exact count.
 In both methods, most of the low-frequency 
\begin_inset Formula $k$
\end_inset

-mers are not reported.
\end_layout

\begin_layout Subsection
One pass method
\begin_inset CommandInset label
LatexCommand label
name "sub:One-pass-method"

\end_inset


\end_layout

\begin_layout Standard
Adding the 
\family sans

\begin_inset Formula $\switch{bf-size}$
\end_inset


\family default
 switch make jellyfish first insert all 
\begin_inset Formula $k$
\end_inset

-mers first into a Bloom filter and only insert into the hash the 
\begin_inset Formula $k$
\end_inset

-mers which have already been seen at least once.
 The argument to 
\begin_inset Formula $\switch{bf-size}$
\end_inset

 should the total number of 
\begin_inset Formula $k$
\end_inset

-mer expected in the data set while the 
\family sans

\begin_inset Formula $\switch{size}$
\end_inset

 argument should be the number of 
\begin_inset Formula $k$
\end_inset

-mers occurring more than once.
 For example:
\end_layout

\begin_layout LyX-Code
jellyfish count -m 25 -s 3G --bf-size 100G -t 16 homo_sapiens.fa
\end_layout

\begin_layout Standard
would be appropriate for counting 
\begin_inset Formula $25$
\end_inset

-mers in human reads at 
\begin_inset Formula $30\times$
\end_inset

 coverage.
 The approximate memory usage is 
\begin_inset Formula $\SI{9}{bits}$
\end_inset

 per 
\begin_inset Formula $k$
\end_inset

-mer in the Bloom filter.
\end_layout

\begin_layout Standard
The count reported for each 
\begin_inset Formula $k$
\end_inset

-mer (by '
\family sans
jellyfish dump
\family default
' or '
\family sans
jellyfish query
\family default
') is one less than the actual count.
 Meaning, the count 
\begin_inset Formula $1$
\end_inset

 
\begin_inset Formula $k$
\end_inset

-mer are not reported, count 
\begin_inset Formula $2$
\end_inset

 
\begin_inset Formula $k$
\end_inset

-mer are reported to have count 
\begin_inset Formula $1$
\end_inset

, etc.
\end_layout

\begin_layout Standard
The drawback of this method is some percentage of the 
\begin_inset Formula $k$
\end_inset

-mer that should not be reported (because they occur only once) are reported.
 This is due to the random nature of the Bloom filter data structure.
 The percentage is 
\begin_inset Formula $<1\%$
\end_inset

 by default and can be changed with the
\begin_inset Formula $\switch{bf-fp}$
\end_inset

 switch.
\end_layout

\begin_layout Subsection
Two pass method
\end_layout

\begin_layout Standard
In the two pass method, first a Bloom counter is created from the reads
 with '
\family sans
jellyfish bc
\family default
'.
 Then this Bloom counter is given to the '
\family sans
jelllyfish count
\family default
' command and only the 
\begin_inset Formula $k$
\end_inset

-mers which have been seen twice in the first pass will be inserted in the
 hash.
 For example, with a human data set similar that in section
\begin_inset space ~
\end_inset


\begin_inset CommandInset ref
LatexCommand ref
reference "sub:One-pass-method"

\end_inset

:
\end_layout

\begin_layout LyX-Code
jellyfish bc -m 25 -s 100G -t 16 -o homo_sapiens.bc homo_sapiens.fa
\end_layout

\begin_layout LyX-Code
jellyfish count -m 25 -s 3G -t 16 --bc homo_sapiens.bc homo_sapiens.fa
\end_layout

\begin_layout Standard
The advantage of this method is that the counts reported for the 
\begin_inset Formula $k$
\end_inset

-mers are all correct.
 Most count 
\begin_inset Formula $1$
\end_inset

 
\begin_inset Formula $k$
\end_inset

-mer are not reported, except for a small percentage (set by the 
\family sans

\begin_inset Formula $\opt f$
\end_inset


\family default
 switch of the 
\family sans
bc
\family default
 subcommand) of them which are reported (correctly with count 
\begin_inset Formula $1$
\end_inset

).
 All other 
\begin_inset Formula $k$
\end_inset

-mers are reported with the correct count.
\end_layout

\begin_layout Standard
The drawback of this method is that it requires to parse the entire reads
 data set twice and the memory usage of the Bloom counter is greater than
 that of the Bloom filter (slightly less than twice as much).
\end_layout

\begin_layout Chapter
FAQ
\end_layout

\begin_layout Section
How to read compressed files (or other format)?
\end_layout

\begin_layout Standard
Jellyfish only reads FASTA or FASTQ formatted input files.
 By reading from pipes, jellyfish can read compressed files, like this:
\end_layout

\begin_layout LyX-Code
zcat *.fastq.gz | jellyfish count /dev/fd/0 ...
\end_layout

\begin_layout Standard
or by using the 
\family sans
'<()
\family default
' 
\family sans
redirection
\family default
 provided by the shell (e.g.
\begin_inset ERT
status open

\begin_layout Plain Layout


\backslash
@
\end_layout

\end_inset

 bash, zsh):
\end_layout

\begin_layout LyX-Code
jellyfish count <(zcat file1.fastq.gz) <(zcat file2.fasta.gz) ...
\end_layout

\begin_layout Section
How to read multiple files at once?
\end_layout

\begin_layout Standard
Often, jellyfish can parse an input sequence file faster than 
\family sans
gzip
\family default
 or 
\family sans
fastq-dump
\family default
 (to parse SRA files) can output the sequence.
 This leads to many threads in jellyfish going partially unused.
 Jellyfish can be instructed to open multiple file at once.
 For example, to read two short read archive files simultaneously:
\end_layout

\begin_layout LyX-Code
jellyfish count -F 2 <(fastq-dump -Z file1.sra) <(fastq-dump -Z file2.sra)
 ...
\end_layout

\begin_layout Standard
Another way is to use 
\begin_inset Quotes eld
\end_inset

generators
\begin_inset Quotes erd
\end_inset

.
 First, create a file containing, one per line, commands to generate sequence.
 Then pass this file to jellyfish and the number of generators to run simultaneo
usly.
 Jellyfish will spawn subprocesses running the commands passed and read
 their standard output for sequence.
 By default, the commands are run using the shell in the SHELL environment
 variable, and this can be changed by the 
\family sans

\begin_inset Formula $\opt S$
\end_inset


\family default
 switch.
 Multiple generators will be run simultaneously as specified by the 
\family sans

\begin_inset Formula $\opt G$
\end_inset


\family default
 switch.
 For example:
\end_layout

\begin_layout LyX-Code
ls *.fasta.gz | xargs -n 1 echo gunzip -c > generators
\end_layout

\begin_layout LyX-Code
jellyfish count -g generators -G 4 ...
\end_layout

\begin_layout Standard
The first command created the command list into the '
\family sans
generators
\family default
' file, each command unzipping one FASTA file in the current directory.
 The second command runs jellyfish with 
\begin_inset Formula $4$
\end_inset

 concurrent generators.
\end_layout

\begin_layout Section
How to reduce the output size?
\end_layout

\begin_layout Standard
The output file was design to be easy to read, but the file generated can
 be rather large.
 By default, a 
\begin_inset Formula $\SI{4}{bytes}$
\end_inset

 counter value is saved for every 
\begin_inset Formula $k$
\end_inset

-mer (i.e.
\begin_inset ERT
status open

\begin_layout Plain Layout


\backslash
@
\end_layout

\end_inset

 a maximum count of over 4 billion).
 Instead, a counter size of 
\begin_inset Formula $\SI{2}{bytes}$
\end_inset

 or 
\begin_inset Formula $\SI{1}{byte}$
\end_inset

 can be used with the switch 
\begin_inset Formula $\switch{out-counter-len}$
\end_inset

, which reduces significantly the output size.
\end_layout

\begin_layout Standard
The count of 
\begin_inset Formula $k$
\end_inset

-mers which cannot be represented with the given number of bytes will have
 a value equal to the maximum value that can be represented.
 Meaning, if the counter field uses 
\begin_inset Formula $\SI{1}{byte}$
\end_inset

, any 
\begin_inset Formula $k$
\end_inset

-mers with count greater or equal to 
\begin_inset Formula $255$
\end_inset

 will be reported of having a count 
\begin_inset Formula $255$
\end_inset

.
\end_layout

\begin_layout Standard
Also, low frequency and high frequency 
\begin_inset Formula $k$
\end_inset

-mers can be skipped using the 
\family sans

\begin_inset Formula $\opt L$
\end_inset


\family default
 and 
\family sans

\begin_inset Formula $\opt U$
\end_inset


\family default
 switches respectively.
 Although it might be more appropriate to filter out the low frequency 
\begin_inset Formula $k$
\end_inset

-mers using Bloom filters, as shown in section
\begin_inset space ~
\end_inset


\begin_inset CommandInset ref
LatexCommand ref
reference "sec:Counting-high-frequency--mers"

\end_inset

.
\end_layout

\begin_layout Section
How much memory is needed?
\end_layout

\begin_layout Standard
The memory needed to count 
\begin_inset Formula $k$
\end_inset

-mers, given the various parameters of the 
\family sans
count 
\family default
subcommand is obtained by using the 
\family sans
mem
\family default
 subcommand.
 It understand all the same switches, but only the 
\begin_inset Formula $\switch{mer-len}$
\end_inset

 (
\begin_inset Formula $\opt m$
\end_inset

), 
\begin_inset Formula $\switch{size}$
\end_inset

 (
\begin_inset Formula $\opt s$
\end_inset

), 
\begin_inset Formula $\switch{counter-len}$
\end_inset

 (
\begin_inset Formula $\opt c$
\end_inset

) and 
\begin_inset Formula $\switch{reprobes}$
\end_inset

 (
\begin_inset Formula $\opt p$
\end_inset

) switches are taken into account.
\end_layout

\begin_layout Standard
For example, for 
\begin_inset Formula $24$
\end_inset

-mers, with an initial hash size of 1 billion, and default parameters otherwise,
 the memory usage is:
\end_layout

\begin_layout LyX-Code
$ jellyfish mem -m 24 -s 1G
\end_layout

\begin_layout LyX-Code
4521043056 (4G)
\end_layout

\begin_layout Standard
Conversely, if the 
\begin_inset Formula $\switch{size}$
\end_inset

 switch is not given by the 
\begin_inset Formula $\switch{mem}$
\end_inset

 switch is, then the maximum initial hash size that would fit in the given
 memory is returned.
 For example, this is the maximum hash size for 
\begin_inset Formula $31$
\end_inset

-mers in 
\begin_inset Formula $\unit[8]{Gb}$
\end_inset

 of RAM:
\end_layout

\begin_layout LyX-Code
$ jellyfish mem -m 31 --mem 8g
\end_layout

\begin_layout LyX-Code
1073741824 (1G)
\end_layout

\begin_layout Chapter
Subcommands
\end_layout

\begin_layout Section

\family sans
histo
\begin_inset CommandInset label
LatexCommand label
name "sec:histo"

\end_inset


\end_layout

\begin_layout Standard
The 
\family sans
histo
\family default
 subcommand outputs the histogram of 
\begin_inset Formula $k$
\end_inset

-mers frequencies.
 The last bin, with value one above the high setting set by the 
\family sans

\begin_inset Formula $\opt h$
\end_inset


\family default
 switch (
\begin_inset Formula $10\,000$
\end_inset

 by default), is a catch all: all 
\begin_inset Formula $k$
\end_inset

-mers with a count greater than the high setting are tallied in that one
 bin.
 If the low setting is set (
\family sans

\begin_inset Formula $\opt l$
\end_inset


\family default
 switch), then the first bin, with value one below the low setting, is also
 similarly a catch all.
\end_layout

\begin_layout Standard
By default, the bins with a zero count are skipped.
 This can be changed with the 
\family sans

\begin_inset Formula $\opt f$
\end_inset


\family default
 switch.
\end_layout

\begin_layout Section

\family sans
dump
\family default

\begin_inset CommandInset label
LatexCommand label
name "sec:dump"

\end_inset


\end_layout

\begin_layout Standard
The 
\family sans
dump
\family default
 subcommand outputs a list of all the 
\begin_inset Formula $k$
\end_inset

-mers in the file associated with their count.
 By default, the output is in FASTA format, where the header line contains
 the count of the 
\begin_inset Formula $k$
\end_inset

-mer and the sequence part is the sequence of the 
\begin_inset Formula $k$
\end_inset

-mer.
 This format has the advantage that the output contains the sequence of
 
\begin_inset Formula $k$
\end_inset

-mers and can be directly fed into another program expecting the very common
 FASTA format.
 A more convenient column format (for human beings) is selected with the
 
\family sans
-c
\family default
 switch.
\end_layout

\begin_layout Standard
Low frequency and high frequency 
\begin_inset Formula $k$
\end_inset

-mers can be skipped with the 
\family sans

\begin_inset Formula $\opt L$
\end_inset


\family default
 and 
\family sans

\begin_inset Formula $\opt U$
\end_inset


\family default
 switches respectively.
\end_layout

\begin_layout Standard
In the output of the 
\family sans
dump
\family default
 subcommand, the 
\begin_inset Formula $k$
\end_inset

-mers are sorted according to the hash function used by Jellyfish.
 The output can be considered to be 
\begin_inset Quotes eld
\end_inset

fairly pseudo-random
\begin_inset Quotes erd
\end_inset

.
 By 
\begin_inset Quotes eld
\end_inset

fairly
\begin_inset Quotes erd
\end_inset

 we mean that NO guarantee is made about the actual randomness of this order,
 it is just good enough for the hash table to work properly.
 And by 
\begin_inset Quotes eld
\end_inset

pseudo-random
\begin_inset Quotes erd
\end_inset

 we mean that the order is actually deterministic: given the same hash function,
 the output will be always the same and two different files generated with
 the same hash function can be merged easily.
\end_layout

\begin_layout Section

\family sans
query
\begin_inset CommandInset label
LatexCommand label
name "sec:query"

\end_inset


\end_layout

\begin_layout Standard
The 
\family sans
query
\family default
 subcommand outputs the 
\begin_inset Formula $k$
\end_inset

-mers and their counts for some subset of 
\begin_inset Formula $k$
\end_inset

-mers.
 It will outputs the counts of all the 
\begin_inset Formula $k$
\end_inset

-mers passed on the command line or of all the 
\begin_inset Formula $k$
\end_inset

-mers in the sequence read from the FASTA or FASTQ formatted file passed
 to the switch 
\family sans

\begin_inset Formula $\opt s$
\end_inset


\family default
 (this switch can be given multiple times).
\end_layout

\begin_layout Section

\family sans
info
\family default

\begin_inset CommandInset label
LatexCommand label
name "sec:info"

\end_inset


\end_layout

\begin_layout Standard
The 
\family sans
info
\family default
 subcommand outputs some information about the jellyfish file and the command
 used to generated it, in which directory and at what time the command was
 run.
 Hopefully, the information given should be enough to rerun jellyfish under
 the same conditions and reproduce the output file.
 In particular, the 
\family sans

\begin_inset Formula $\opt c$
\end_inset


\family default
 switch outputs the command, properly escaped and ready to run in a shell.
\end_layout

\begin_layout Standard
The header is saved in JSON format and contains more information than is
 written by the default.
 The full header in JSON format can be written out using the 
\family sans
-j
\family default
 switch.
\end_layout

\begin_layout Section

\family sans
merge
\family default

\begin_inset CommandInset label
LatexCommand label
name "sec:merge"

\end_inset


\end_layout

\begin_layout Standard
The 
\family sans
merge
\family default
 subcommand is a little direct use with version version 2 of jellyfish.
 When intermediary files were written to disk, because not all 
\begin_inset Formula $k$
\end_inset

-mers would fit in memory, they can be merged into one file containing the
 final result with the 
\family sans
merge
\family default
 subcommand.
 The 
\family sans
count
\family default
 will merge intermediary files automatically as needed.
\end_layout

\begin_layout Section
stats 
\begin_inset CommandInset label
LatexCommand label
name "sec:stats"

\end_inset


\end_layout

\begin_layout Standard
The 
\family sans
stats
\family default
 subcommand computes some statistics about the mers.
 Although these statistics could be computed from the histogram, it provides
 quick summary information.
 The fields are:
\end_layout

\begin_layout Description
Unique The number of 
\begin_inset Formula $k$
\end_inset

-mer occuring exactly once
\end_layout

\begin_layout Description
Distinct The number of 
\begin_inset Formula $k$
\end_inset

-mers, ignoring their multiplicity (i.e.
\begin_inset ERT
status open

\begin_layout Plain Layout


\backslash
@
\end_layout

\end_inset

 the cardinality of the set of 
\begin_inset Formula $k$
\end_inset

-mers)
\end_layout

\begin_layout Description
Total The number of 
\begin_inset Formula $k$
\end_inset

-mers with multiplicity (i.e.
\begin_inset ERT
status open

\begin_layout Plain Layout


\backslash
@
\end_layout

\end_inset

 the sum of the number of occurence of all the mers)
\end_layout

\begin_layout Description
Max_count The maximum of the number of occurences
\end_layout

\begin_layout Section
mem 
\begin_inset CommandInset label
LatexCommand label
name "sec:mem"

\end_inset


\end_layout

\begin_layout Standard
The 
\family sans
mem
\family default
 subcommand shows how much memory a count subcommand will need or conversely
 how large of a hash size will fit in a given amount of memory.
\end_layout

\begin_layout Section
cite
\begin_inset CommandInset label
LatexCommand label
name "sec:cite"

\end_inset


\end_layout

\begin_layout Standard
The 
\family sans
cite
\family default
 subcommand prints the citation for the jellyfish paper.
 With the 
\begin_inset Formula $\opt b$
\end_inset

, it is formatted in Bibtex format.
 How convenient!
\end_layout

\end_body
\end_document