primer3 release 0.6
Copyright (c) 1996,1997
Whitehead Institute for Biomedical Research. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
1. Redistributions must reproduce the above copyright notice, this
list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution. Redistributions of
source code must also reproduce this information in the source code itself.
2. If the program is modified, redistributions must include a notice
(in the same places as above) indicating that the redistributed program is
not identical to the version distributed by Whitehead Institute.
3. All advertising materials mentioning features or use of this
software must display the following acknowledgment:
This product includes software developed by the
Whitehead Institute for Biomedical Research.
4. The name of the Whitehead Institute may not be used to endorse or
promote products derived from this software without specific prior written
permission.
We also request that use of this software be cited in publications as
Steve Rozen, Helen J. Skaletsky (1996,1997)
Primer3. Code available at
http://www-genome.wi.mit.edu/genome_software/other/primer3.html
THIS SOFTWARE IS PROVIDED BY THE WHITEHEAD INSTITUTE ``AS IS'' AND ANY
EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE WHITEHEAD INSTITUTE BE LIABLE FOR ANY
DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
SUCH DAMAGE.
INTRODUCTION
------------
Primer3 is a complete rewrite of the original PRIMER program
(Primer 0.5), written by Steve Lincoln, Mark Daly, and Eric
Lander. See DIFFERENCES FROM EARLIER VERSIONS for a discussion
of how Primer3 differs from its predecessors, Primer 0.5 and
Primer v2.
Primer3 picks primers for PCR reactions, considering as criteria:
o oligonucleotide melting temperature, size, GC content,
and primer-dimer possibilities,
o PCR product size,
o positional constraints within the source sequence, and
o miscellaneous other constraints.
All of these criteria are user-specifiable as constraints, and
some are specifiable as terms in an objective function that
characterizes an optimal primer pair.
The Whitehead/MIT Center for Genome Research offers web-based
front end to Primer3 at
http://www.genome.wi.mit.edu/cgi-bin/primer/primer3.cgi.
INSTALLATION INSTRUCTIONS
-------------------------
Go to the src directory of the distribution.
If you do not use gcc, modify the makefile to
use your (ANSI) C compiler and appropriate
compile and link flags.
$ make primer3
# Warnings about i0, j0, I, J, and score being possibly used
# unset in dpal.c are normal and harmless.
# Warnings about pr_release being unused are harmless.
$ cd ../test
# If you do not have perl5 in /usr/local/bin/perl5
# modify the first line of primer_test.pl to point
# to your path for perl 5.
$ primer_test.pl
# You should not see 'FAILED' during the tests.
SYSTEM REQUIREMENTS
-------------------
Primer3 has been successfully installed and tested on the
following systems
o Sparc running SunOS 4.1 (gcc 2.7.0)
o Alpha running DEC Unix 3.2 (gcc 2.7.0 and DEC cc)
o Pentium running Linux 1.2 (Red Hat) (gcc 2.7.0)
Primer3 may well compile and run on other POSIX architectures with
ANSI C compilers. Please contact the authors with portability
suggestions.
INPUT AND OUTPUT CONVENTIONS
----------------------------
By default, Primer3 accepts input and produces output in
Boulder-io format, a text-based input/output format used as a
program-to-program data interchange format in many information
systems at the Whitehead/MIT Center for Genome Research. See
http://www.genome.wi.mit.edu and follow the links to "Genome
Center Software" for more information. When run with the
-format_output command-line flag, Primer3 prints a more
user-oriented report for each sequence. Additional command-line
flags include -2x_compat (which causes Primer3 to print its
output using Primer v2 compatible tag names), and -strict_tags
(both discussed below). Primer3 exits with 0 status if it
operates correctly. See EXIT STATUS CODES below for additional
information.
The syntax of the version of Boulder-io recognized by Primer3 is
as follows:
o Input consists of a sequence of RECORDs.
o A RECORD consists of a sequence of (TAG,VALUE) pairs, each terminated
by a newline character (\n). A RECORD is terminated by '='
appearing by itself on a line.
o A (TAG,VALUE) pair has the following requirements:
o the TAG must be immediately (without spaces)
followed by '='.
o the pair must be terminated by a newline character.
An example of a legal (TAG,VALUE) pair is
PRIMER_SEQUENCE_ID=my_marker
and an example of a BOULDER-IO record is
PRIMER_SEQUENCE_ID=test1
SEQUENCE=GACTGATCGATGCTAGCTACGATCGATCGATGCATGCTAGCTAGCTAGCTGCTAGC
=
Many records can be sent, one after another. Below is an example
of three different records which might be passed through a
boulder-io stream:
PRIMER_SEQUENCE_ID=test1
SEQUENCE=GACTGATCGATGCTAGCTACGATCGATCGATGCATGCTAGCTAGCTAGCTGCTAGC
=
PRIMER_SEQUENCE_ID=test2
SEQUENCE=CATCATCATCATCGATGCTAGCATCNNACGTACGANCANATGCATCGATCGT
=
PRIMER_SEQUENCE_ID=test3
SEQUENCE=NACGTAGCTAGCATGCACNACTCGACNACGATGCACNACAGCTGCATCGATGC
=
Primer3 reads boulder-io on stdin and echos its input and returns
results in boulder-io format on stdout. Primer3 indicates many
user-correctable errors by a value in the PRIMER_ERROR tag (see
below) and indicates other errors, including system configuration
errors, resource errors (such out-of-memory errors), and detected
programming errors by a message on stderr and a non-zero exit
status.
Below is the list of input tags that Primer3 recognizes.
Primer3 echos and ignores any tags it does not recognize, unless
the -strict_tags flag is set on the command line, in which case
Primer3 prints an error in the PRIMER_ERROR output tag (see
below), and prints additional information on stdout; this option
can be useful for debugging systems that incorporate primer.
Except for tags with the type "interval list" each tag is allowed
only ONCE in any given input record. This restriction is not
systematically checked in this beta release: use care.
There are 2 major classes of input tags. "Sequence" input tags
describe a particular input sequence to Primer3, and are reset
after every boulder record. "Global" input tags describe the
general parameters that Primer3 should use in its searches, and
the values of these tags persist between input boulder records
until or unless they are explicitly reset. Errors in "Sequence"
input tags invalidate the current record, but Primer3 will
continue to process additional records. Errors in "Global" input
tags are fatal because they invalidate the basic conditions under
which primers are being picked.
"Sequence" Input Tags
---------------------
PRIMER_SEQUENCE_ID (string, optional)
(MARKER_NAME is a deprecated synonym maintained for v2
compatibility.)
An identifier that is reproduced in the output to enable users to
identify the source of the chosen primers.
This tag must be present if PRIMER_FILE_FLAG is non-zero.
SEQUENCE (nucleotide sequence, REQUIRED)
The sequence from which to choose primers. The sequence
must be presented 5' -> 3' (see the discussion of the
PRIMER_SELF_END argument). The bases may be upper or lower case.
No newlines should be inserted into the sequence, because the
Boulder-IO parser will assume that a line ends at a newline.
INCLUDED_REGION (interval, optional)
A sub-region of the given sequence in which to pick primers. For
example, often the first dozen or so bases of a sequence are
vector, and should be excluded from consideration. The value for
this parameter has the form
,
where is the 0-based index of the first base to consider,
and is the number of subsequent bases in the
primer-picking region.
TARGET (interval list, default empty)
If one or more Targets is specified then a legal primer pair must
flank at least one of them. A Target might be a simple sequence
repeat site (for example a CA repeat) or a single-base-pair
polymorphism. The value should be a space-separated list of
,
pairs where is the (0-based) index of the first base of a
Target, and is its length.
For backward compatibility Primer3 accepts (but ignores)
a trailing , for each element of this argument.
EXCLUDED_REGION (interval list, default empty)
Primer oligos may not overlap any region specified in this tag.
The associated value must be a space-separated list of
,
pairs where is the (0-based) index of the first base of
the excluded region, and is its length. This tag is
useful for tasks such as excluding regions of low sequence
quality or for excluding regions containing repetitive elements
such as ALUs or LINEs.
PRIMER_COMMENT (string, optional)
The value of this tag is ignored.
COMMENT (string, optional)
Deprecated synonym for PRIMER_COMMENT.
PRIMER_SEQUENCE_QUALITY (quality list, default empty)
A list of space separated integers. There must be exactly one
integer for each base in SEQUENCE if this argument is non-empty.
High numbers indicate high confidence in the base call at that
position and low numbers indicate low confidence in the base
call at that position.
PRIMER_LEFT_INPUT (nucleotide sequence, default empty)
The sequence of a left primer to check and around which to design
right primers and optional internal oligos. Must be a substring
of SEQUENCE.
PRIMER_RIGHT_INPUT (nucleotide sequence, default empty)
The sequence of a right primer to check and around which to
design left primers and optional internal oligos. Must be a
substring of the reverse strand of SEQUENCE.
"Global" Input Tags
-------------------
PRIMER_MISPRIMING_LIBRARY (string, optional)
The name of a file containing a nucleotide sequence library of
sequences to avoid amplifying (for example repetitive sequences, or
possibly the sequences of genes in a gene family that should
not be amplified.) The file must be in (a slightly restricted)
FASTA format (W. B. Pearson and D.J. Lipman, PNAS 85:8 pp
2444-2448 [1988]); we briefly discuss the organization of this
file below. If this parameter is specified then Primer3 locally
aligns each candidate primer against each library sequence and
rejects those primers for which the local alignment score times a
specified weight (see below) exceeds PRIMER_MAX_MISPRIMING.
(The maximum value of the weight is arbitrarily set to 100.0.)
Each sequence entry in the FASTA-format file must begin with an
"id line" that starts with '>'. The contents of the id line is
"slightly restricted" in that Primer3 parses everything after any
optional asterisk ('*') as a floating point number to use as the
weight mentioned above. If the id line contains no asterisk then
the weight defaults to 1.0. The alignment scoring system used is
the same as for calculating complementarity among oligos (e.g.
PRIMER_SELF_ANY). The remainder of an entry contains the
sequence as lines following the id line up until a line starting
with '>' or the end of the file. Whitespace and newlines are
ignored. Characters 'A', 'T', 'G', 'C', 'a', 't', 'g', 'c' are
retained and any other character is converted to 'N' (with the
consequence that any IUB / IUPAC codes for ambiguous bases are
converted to 'N'). There are no restrictions on line length.
An empty value for this parameter indicates that no repeat
library should be used.
PRIMER_MAX_MISPRIMING (decimal,9999.99, default 12.00)
The maximum allowed weighted similarity with any sequence in
PRIMER_MISPRIMING_LIBRARY.
PRIMER_PAIR_MAX_MISPRIMING (decimal,9999.99, default 24.00)
The maximum allowed sum of weighted similarities of a primer pair
(one similarity for each primer) with any single sequence in
PRIMER_MISPRIMING_LIBRARY.
PRIMER_EXPLAIN_FLAG (boolean, default 0)
If this flag is non-0, produce PRIMER_LEFT_EXPLAIN,
PRIMER_RIGHT_EXPLAIN, and PRIMER_INTERNAL_OLIGO_EXPLAIN output
tags, which are intended to provide information on the number of
oligos and primer pairs that Primer3 examined, and statistics on
the number discarded for various reasons. If -format_output is
set similar information is produced in the user-oriented output.
PRIMER_PRODUCT_SIZE_RANGE (size range list, default 100-300)
The associated values specify the lengths of the product that the
user wants the primers to create, and is a space separated list
of elements of the form
-
where an - pair is a legal range of lengths for the
product. For example, if one wants PCR products to be between
100 to 150 bases (inclusive) then one would set this parameter to
100-150. If one desires PCR products in either the range from
100 to 150 bases or in the range from 200 to 250 bases then one
would set this parameter to 100-150 200-250.
Primer3 favors ranges to the left side of the parameter string.
Primer3 will return legal primers pairs in the first range
regardless the value of the objective function for these pairs.
Only if there are an insufficient number of primers in the first
range will Primer3 return primers in a subsequent range.
PRIMER_PICK_INTERNAL_OLIGO (boolean, default 0)
If the associated value is non-0, then Primer3 will attempt to
pick an internal oligo. Briefly, an "internal oligo" is intended
to be used as a hybridization probe to detect the PCR product
after amplification.
PRIMER_GC_CLAMP (int, default 0)
Require the specified number of consecutive Gs and Cs at the 3'
end of both the left and right primer. (This parameter has no
effect on the internal oligo if one is requested.)
PRIMER_OPT_SIZE (int, default 20)
Optimum length (in bases) of a primer oligo. Primer3 will attempt
to pick primers close to this length.
PRIMER_DEFAULT_SIZE (int, default 20)
A deprecated synonym for PRIMER_OPT_SIZE, maintained for v2
compatibility.
PRIMER_MIN_SIZE (int, default 18)
Minimum acceptable length of a primer.
PRIMER_MAX_SIZE (int, default 27)
Maximum acceptable length (in bases) of a primer. Currently this
parameter cannot be larger than 35. This limit is governed by
maximum oligo size for which Primer3's melting-temperature is
valid.
PRIMER_OPT_TM (float, default 60.0C)
Optimum melting temperature(Celsius) for a primer oligo. Primer3
will try to pick primers with melting temperatures are close to
this temperature. The oligo melting temperature formula in
Primer3 is that given in Rychlik, Spencer and Rhoads, Nucleic
Acids Research, vol 18, num 12, pp 6409-6412 and Breslauer,
Frank, Bloeker and Marky, Proc. Natl. Acad. Sci. USA, vol 83, pp
3746-3750. Please refer to the former paper for background
discussion.
PRIMER_MIN_TM (float, default 57.0C)
Minimum acceptable melting temperature(Celsius) for a primer
oligo.
PRIMER_MAX_TM (float, default 63.0C)
Maximum acceptable melting temperature(Celsius) for a primer
oligo.
PRIMER_MAX_DIFF_TM (float, default 100.0C)
Maximum acceptable (unsigned) difference between the melting
temperatures of the left and right primers.
PRIMER_MIN_GC (float, default 20.0%)
Minimum allowable percentage of Gs and Cs in any primer.
PRIMER_MAX_GC (float, default 80.0%)
Maximum allowable percentage of Gs and Cs in any primer generated
by Primer.
PRIMER_SALT_CONC (float, default 50.0 mM)
The millimolar concentration of salt (usually KCl) in the PCR.
Primer3 uses this argument to calculate oligo melting
temperatures.
PRIMER_DNA_CONC (float, default 50.0 nM)
The nanomolar concentration of annealing oligos in the PCR.
Primer3 uses this argument to calculate oligo melting
temperatures. The default (50nM) works well with the standard
protocol used at the Whitehead/MIT Center for Genome
Research--0.5 microliters of 20 micromolar concentration for each
primer oligo in a 20 microliter reaction with 10 nanograms
template, 0.025 units/microliter Taq polymerase in 0.1 mM each
dNTP, 1.5mM MgCl2, 50mM KCl, 10mM Tris-HCL (pH 9.3) using 35
cycles with an annealing temperature of 56 degrees Celsius. This
parameter corresponds to 'c' in Rychlik, Spencer and Rhoads'
equation (ii) (Nucleic Acids Research, vol 18, num 12) where a
suitable value (for a lower initial concentration of template) is
"empirically determined". The value of this parameter is less
than the actual concentration of oligos in the reaction because
it is the concentration of annealing oligos, which in turn
depends on the amount of template (including PCR product) in a
given cycle. This concentration increases a great deal during a
PCR; fortunately PCR seems quite robust for a variety of oligo
melting temperatures.
See ADVICE FOR PICKING PRIMERS.
PRIMER_NUM_NS_ACCEPTED (int, default 0)
Maximum number of unknown bases (N) allowable in any primer.
PRIMER_SELF_ANY (decimal,9999.99, default 8.00)
The maximum allowable local alignment score when testing a single
primer for (local) self-complementarity and the maximum allowable
local alignment score when testing for complementarity between
left and right primers. Local self-complementarity is taken to
predict the tendency of primers to anneal to each other without
necessarily causing self-priming in the PCR. The scoring system
gives 1.00 for complementary bases, -0.25 for a match of any base
(or N) with an N, -1.00 for a mismatch, and -2.00 for a gap.
Only single-base-pair gaps are allowed. For example, the
alignment
5' ATCGNA 3'
|| | |
3' TA-CGT 5'
is allowed (and yields a score of 1.75), but the alignment
5' ATCCGNA 3'
|| | |
3' TA--CGT 5'
is not considered. Scores are non-negative, and a score of 0.00
indicates that there is no reasonable local alignment between two
oligos.
PRIMER_SELF_END (decimal 9999.99, default 3.00)
The maximum allowable 3'-anchored global alignment score when
testing a single primer for self-complementarity, and the maximum
allowable 3'-anchored global alignment score when testing for
complementarity between left and right primers. The 3'-anchored
global alignment score is taken to predict the likelihood of
PCR-priming primer-dimers, for example
5' ATGCCCTAGCTTCCGGATG 3'
||| |||||
3' AAGTCCTACATTTAGCCTAGT 5'
or
5` AGGCTATGGGCCTCGCGA 3'
||||||
3' AGCGCTCCGGGTATCGGA 5'
The scoring system is as for the Maximum Complementarity
argument. In the examples above the scores are 7.00 and 6.00
respectively. Scores are non-negative, and a score of 0.00
indicates that there is no reasonable 3'-anchored global
alignment between two oligos. In order to estimate 3'-anchored
global alignments for candidate primers and primer pairs, Primer
assumes that the sequence from which to choose primers is
presented 5'->3'. It is nonsensical to provide a larger value
for this parameter than for the Maximum (local) Complementarity
parameter because the score of a local alignment will always be at
least as great as the score of a global alignment.
PRIMER_DEFAULT_PRODUCT (size range list, default 100-300)
A deprecated synonym for PRIMER_PRODUCT_SIZE_RANGE, maintained
for v2 compatibility.
PRIMER_FILE_FLAG (boolean, default 0)
If the associated value is non-0, then Primer3 creates two output
files for each input SEQUENCE. File .for lists all
acceptable left primers for , and .rev
lists all acceptable right primers for , where
is the value of the PRIMER_SEQUENCE_ID tag (which
must be supplied). In addition, if the input tag
PRIMER_PICK_INTERNAL_OLIGO is non-0, Primer3 produces a file
.int, which lists all acceptable internal oligos.
PRIMER_MAX_POLY_X (int, default 5)
The maximum allowable length of a mononucleotide repeat,
for example AAAAAA.
PRIMER_LIBERAL_BASE (boolean, default 0)
This parameter provides a quick-and-dirty way to get Primer3 to
accept IUB / IUPAC codes for ambiguous bases (i.e. by changing
all unrecognized bases to N). If you wish to include an
ambiguous
base in an oligo, you must set PRIMER_NUM_NS_ACCEPTED to a
non-0 value.
Perhaps '-' and '* ' should be squeezed out rather than changed
to 'N', but currently they simply get converted to N's. The authors
invite user comments.
PRIMER_NUM_RETURN (int, default 5)
The maximum number of primer pairs to return. Primer pairs
returned are sorted by their "quality", in other words by the
value of the objective function (where a lower number indicates a
better primer pair). Caution: setting this parameter to a large
value will increase running time.
PRIMER_FIRST_BASE_INDEX (int, default 0)
This parameter is the index of the first base in the input
sequence. For input and output using 1-based indexing (such as
that used in GenBank and to which many users are accustomed) set
this parameter to 1. For input and output using 0-based indexing
set this parameter to 0. (This parameter also affects the
indexes in the contents of the files produced when the primer
file flag is set.)
PRIMER_MIN_QUALITY (int, default 0)
The minimum sequence quality (as specified by
PRIMER_SEQUENCE_QUALITY) allowed within a primer.
PRIMER_MIN_END_QUALITY (int, default 0)
The minimum sequence quality (as specified by
PRIMER_SEQUENCE_QUALITY) allowed within the 5' pentamer of a
primer.
PRIMER_QUALITY_RANGE_MIN (int, default 0)
The minimum legal sequence quality (used for error checking
of PRIMER_MIN_QUALITY and PRIMER_MIN_END_QUALITY).
PRIMER_QUALITY_RANGE_MAX (int, default 100)
The maximum legal sequence quality (used for error checking
of PRIMER_MIN_QUALITY and PRIMER_MIN_END_QUALITY).
PRIMER_INSIDE_PENALTY (float, default -1.0)
This experimental parameter might not be maintained in this form
in the next release. Non-default values valid only for sequences
with 0 or 1 target regions. If the primer is part of a pair that
spans a target and overlaps the target, then multiply this value
times the number of nucleotide positions by which the primer
overlaps the (unique) target to get the 'position penalty'. The
effect of this parameter is to allow Primer3 to include overlap
with the target as a term in the objective function.
PRIMER_OUTSIDE_PENALTY (float, default 0.0)
This experimental parameter might not be maintained in this form
in the next release. Non-default values valid only for sequences
with 0 or 1 target regions. If the primer is part of a pair that
spans a target and does not overlap the target, then multiply
this value times the number of nucleotide positions from the 3'
end to the (unique) target to get the 'position penalty'.
The effect of this parameter is to allow Primer3 to include
nearness to the target as a term in the objective function.
PRIMER_MAX_END_STABILITY (float 999.9999, default 100.0)
The maximum stability for the five 3' bases of a left or right
primer. Bigger numbers mean more stable 3' ends. The value is
the maximum delta G for duplex disruption for the five 3' bases
as calculated using the nearest neighbor parameters published in
Breslauer, Frank, Bloeker and Marky, Proc. Natl. Acad. Sci. USA,
vol 83, pp 3746-3750. Primer3 uses a completely permissive
default value for backward compatibility (which we may change in
the next release). Rychlik recommends a maximum value of 9
(Wojciech Rychlik, "Selection of Primers for Polymerase Chain
Reaction" in BA White, Ed., "Methods in Molecular Biology,
Vol. 15: PCR Protocols: Current Methods and Applications", 1993,
pp 31-40, Humana Press, Totowa NJ).
Like the arguments governing PCR primer selection, the input tags
governing internal oligo selection are divided into sequence
input tags and global input tags, with for former being
automatically reset after each input record, and the latter
persisting until explicitly reset.
Because the laboratory detection step using internal oligos
is independent of the PCR amplification procedure,
internal oligo tags have defaults that are independent
of the parameters that govern the selection of PCR primers.
For example, the melting temperature of an oligo
used for hybridization might be considerably lower
than that used as a PCR primer.
Internal Oligo "Sequence" Input Tags
------------------------------------
PRIMER_INTERNAL_OLIGO_EXCLUDED_REGION (interval list, default empty)
Middle oligos may not overlap any region specified by this tag.
The associated value must be a space-separated list of
,
pairs, where is the (0-based) index of the first base of
an excluded region, and is its length. Often one would
make Target regions excluded regions for internal oligos.
PRIMER_INTERNAL_OLIGO_INPUT (nucleotide sequence, default empty)
The sequence of an internal oligo to check and around which to
design left and right primers. Must be a substring of SEQUENCE.
Internal Oligo "Global" Input Tags
----------------------------------
These tags are analogous to the global input tags (those
governing primer oligos) discussed above. The exception is
PRIMER_INTERNAL_OLIGO_SELF_END which is meaningless when applied
to internal oligos used for hybridization-based detection, since
primer-dimer will not occur. We recommend that
PRIMER_INTERNAL_OLIGO_SELF_END be set at least as high as
PRIMER_INTERNAL_OLIGO_SELF_ANY.
PRIMER_INTERNAL_OLIGO_OPT_SIZE (int, default 20)
PRIMER_INTERNAL_OLIGO_MIN_SIZE (int, default 18)
PRIMER_INTERNAL_OLIGO_MAX_SIZE (int, default 27)
PRIMER_INTERNAL_OLIGO_OPT_TM (float, default 60.0 degrees C)
PRIMER_INTERNAL_OLIGO_MIN_TM (float, default 57.0 degrees C)
PRIMER_INTERNAL_OLIGO_MAX_TM (float, default 63.0 degrees C)
PRIMER_INTERNAL_OLIGO_MIN_GC (float, default 20.0%)
PRIMER_INTERNAL_OLIGO_MAX_GC (float, default 80.0%)
PRIMER_INTERNAL_OLIGO_SALT_CONC (float, default 50.0 mM)
PRIMER_INTERNAL_OLIGO_DNA_CONC (float, default 50.0 nM)
PRIMER_INTERNAL_OLIGO_SELF_ANY (decimal 9999.99, default 12.00)
PRIMER_INTERNAL_OLIGO_MAX_POLY_X (int, default 5)
PRIMER_INTERNAL_OLIGO_SELF_END (decimal 9999.99, default 12.00)
PRIMER_INTERNAL_OLIGO_MISHYB_LIBRARY (string, optional)
Similar to PRIMER_MISPRIMING_LIBRARY, except that the event we
seek to avoid is hybridization of the internal oligo to sequences
in this library rather than priming from them.
PRIMER_INTERNAL_OLIGO_MAX_MISHYB (decimal,9999.99, default 12.00)
Similar to PRIMER_MAX_MISPRIMING except that this parameter applies
to the similarity of candidate internal oligos to the library
specified in PRIMER_INTERNAL_OLIGO_MISHYB_LIBRARY.
PRIMER_INTERNAL_OLIGO_MIN_QUALITY (int, default 0)
(Note that there is no PRIMER_INTERNAL_OLIGO_MIN_END_QUALITY.)
AN EXAMPLE
----------
One might be interested in performing PCR on an STS with a CA
repeat in the middle of it. Primers need to be chosen based on
the criteria of the experiment.
We need to come up with a boulder-io record to send to Primer3 via
stdin. There are lots of ways to accomplish this. We could save
the record into a text file called 'input', and then type the
UNIX command 'primer3 < input'. Or, we could simply type the
command 'primer3', which would start primer with no input. Then,
we could type in the record line by line, terminating it with the
'=' character, and then inserting an EOF by holding down the
'ctrl' button while pressing 'd'. In any case, we assume that you
will be able to get this record into Primer3's stdin. See the man
page for your shell if you are having problems with this.
Let's look at the input record itself:
PRIMER_SEQUENCE_ID=example
SEQUENCE=GTAGTCAGTAGACNATGACNACTGACGATGCAGACNACACACACACACACAGCACACAGGTATTAGTGGGCCATTCGATCCCGACCCAAATCGATAGCTACGATGACG
TARGET=37,21
PRIMER_OPT_SIZE=18
PRIMER_MIN_SIZE=15
PRIMER_MAX_SIZE=21
PRIMER_NUM_NS_ACCEPTED=1
PRIMER_PRODUCT_SIZE_RANGE=75-100
PRIMER_FILE_FLAG=1
PRIMER_PICK_INTERNAL_OLIGO=1
PRIMER_INTERNAL_OLIGO_EXCLUDED_REGION=37,21
PRIMER_EXPLAIN_FLAG=1
=
A breakdown of the reasoning behind each of the TAG=VALUE pairs
is below:
PRIMER_SEQUENCE_ID=example
The main intent of this tag is to provide an identifier for the
sequence that is meaningful to the user, for example when Primer3
processes multiple records, and by default this tag is optional.
However, this tag is _required_ when PRIMER_FILE_FLAG is non-0
Because it provides the names of the files that contain lists
of oligos that Primer3 considered.
SEQUENCE=GTAGTCAGTAGACNATGACNACTGACGATGCAGACNACACACACACACACAGCACACAGGTATTAGTGGGCCATTCGATCCCGACCCAAATCGATAGCTACGATGACG
The SEQUENCE tag is of ultimate importance. Without it, Primer3
has no idea what to do. This sequence is 92 bases long. Note that
there is no newline until the sequence terminates completely.
TARGET=37,21
There is a simple sequence repeat in our sequence, which starts
at base 37, and has a length of 21 bases. We want Primer3 to
choose primers which flank the repeat site, so we let Primer3 know
that we consider this site to be important.
PRIMER_OPT_SIZE=18
Since our sequence length is rather small (only 92 bases
long), we lower the PRIMER_OPT_SIZE from 20 to 18. It's
more likely that Primer3 will succeed if it shoots for smaller
primers with such a small sequence.
PRIMER_MIN_SIZE=15
PRIMER_MAX_SIZE=21
With the lowering of optimal primer size, it's good to lower
the minimum and maximum sizes as well.
PRIMER_NUM_NS_ACCEPTED=1
Again, since we've got such a small sequence with a
non-negligible amount of unknown bases (N's) in it, let's make
Primer3's job easier by allowing it to pick primers that have
at most 1 unknown base.
PRIMER_PRODUCT_SIZE_RANGE=75-100
We reduce the product size range from the default of 100-300
because our source sequence is only 108 base pairs long. If we
insisted on a product size of 100 base pairs Primer3 would have
few possibilities to choose from.
PRIMER_FILE_FLAG=1
Since we've got such a small sequence, Primer might fail to
pick primers. We want to get the list of primers it
considered, then, so that we might manually pick primers
ourselves if Primer fails to do so. Setting this flag to 1
will force Primer to output the primers it considered to a
forward_primer and a reverse_primer output file.
PRIMER_PICK_INTERNAL_OLIGO=1
We want to see if Primer v2.3 can pick an internal oligo for
the sequence, so we set this flag to 1 (true).
PRIMER_INTERNAL_OLIGO_EXCLUDED_REGION=37,21
Normally CA-repeats make poor hybridization probes (because they
not specific enough). Therefor we exclude the CA repeat (which
is the TARGET) from consideration for the middle oligo.
PRIMER_EXPLAIN_FLAG=1
We want to see statistics about the oligos and oligo triples
(left primer, internal oligo, right primer) that Primer3
examined.
=
The '=' character terminates the record.
Tere are some boulderio tags that we never even
specified. (INCLUDED_REGION, EXCLUDED_REGION, et al.), which is
perfectly legal. For the tags with default values, those
defaults will be used in the analysis. For the tags with NO
default values (like TARGET, for instance), the functionality
requested by the those tags will simply be absent. It's not the
case that we need to surround a simple sequence repeat every time
we want to pick primers!
OUTPUT TAGS
-----------
For each boulderio record passed into primer3 via stdin, exactly
one boulderio record comes out of primer3 on stdout. These output
records contain everything that the input record contains, plus a
subset of the following tag/value pairs. Unless noted by (*),
each tag appears for each primer pair returned. The first
version is PRIMER_{LEFT,RIGHT,INTERNAL_OLIGO,PAIR}_.
Tags of additional primers chosen are of the form
PRIMER_{LEFT,RIGHT,INTERNAL_OLIGO,PAIR}__. where
is an integer from 1 to n, where n is at most the value of
PRIMER_NUM_RETURN.
In the descriptions below, 'i,n' represents a start/length pair,
's' represents a string, x represents an arbitrary integer, and f
represents a float.
PRIMER_ERROR=s (*)
s describes user-correctible errors detected in the input
(separated by semicolons). This tag is absent if there are no
errors.
PRIMER_LEFT=i,n
(FORWARD_PRIMER if -v2_compat is set)
The selected left primer (the primer to the left in the input
sequence). i is the 0-based index of the start base of the
primer, and n is t its length.
PRIMER_RIGHT=i,n
(REVERSE_PRIMER if -v2_compat is set)
The selected right primer (the primer to the right in the input
sequence). i is the 0-based index of the last base of the
primer, and n is its length.
PRIMER_INTERNAL_OLIGO=i,n
(MIDDLE_OLIGO if -v2_compat is set)
The selected internal oligo. Primer3 outputs this tag if
PRIMER_PICK_INTERNAL_OLIGO was non-0. If primer3 fails to pick a
middle oligo upon request, this tag will not be output. i is the
0-based index of start base of the internal oligo, and n is its
length.
PRIMER_PRODUCT_SIZE=x
(PRODUCT_SIZE if -v2_compat is set)
x is the product size of the PCR product.
PRIMER_{LEFT,RIGHT,INTERNAL_OLIGO}_EXPLAIN=s (*)
s is a (more or less) self-documenting string containing
statistics on the possiblities that primer3 considered in
selecting a single oligo. For example
PRIMER_LEFT_EXPLAIN=considered 62, too many Ns 53, ok 9
PRIMER_RIGHT_EXPLAIN=considered 62, too many Ns 53, ok 9
PRIMER_INTERNAL_OLIGO_EXPLAIN=considered 87, too many Ns 39, overlap excluded region 40, ok 8
All the categories are exclusive, except the 'considered' category.
PRIMER_PAIR_EXPLAIN=s (*)
s is a self-documenting string containing statistics on picking a
primer pair (plus internal oligo if requested). For exaple
PRIMER_PAIR_EXPLAIN=considered 81, unacceptable product size 49, no internal oligo 32, ok 0
All the categories are exclusive, except the 'considered' category.
PRIMER_PAIR_QUALITY=f
The value of the objective function for this pair (lower is better).
PRIMER_{LEFT,RIGHT,INTERNAL_OLIGO}_SEQUENCE=s
The actual sequence of the oligo. The sequence of left primer and
internal oligo is presented 5' -> 3' on the same strand as the
input SEQUENCE (which must be presented 5' -> 3'). The sequence
of the right primer is presented 5' -> 3' on the opposite strand
from the input SEQUENCE.
PRIMER_{LEFT,RIGHT,INTERNAL_OLIGO}_TM=f
The melting TM for the selected oligo.
PRIMER_{LEFT,RIGHT,INTERNAL_OLIGO}_SELF_ANY=f
PRIMER_{LEFT,RIGHT,INTERNAL_OLIGO}_SELF_END=f
The self-complementarity measures for the selected oligo.
PRIMER_PAIR_COMPL_ANY=f
PRIMER_PAIR_COMPL_END=f
The inter-pair complementarity measures for the selected left and
right primer
PRIMER_WARNING=s (*)
s lists warnings generated by primer (separated by semicolons);
this tag is absent if there are no warnings
PRIMER_{LEFT,RIGHT,PAIR}_MISPRIMING_SCORE=f, s
f is the maximum mispriming score for the right primer
against any sequence in the given PRIMER_MISPRIMING_LIBRARY;
s is the id of corresponding library sequence.
PRIMER_PAIR_MISPRIMING_SCORE is the maximum sum of
mispriming scores in any single library sequence (perhaps a
more reasonable estimator of the likelihood of mispriming).
PRIMER_INTERNAL_OLIGO_MISHYB_SCORE=f, s
f is the maximum mishybridization score for the right primer
against any sequence in the given
PRIMER_INTERNAL_OLIGO_MISHYB_LIBRARY; s is the id of
corresponding library sequence.
PRIMER_{LEFT,RIGHT,INTERNAL_OLIGO}_MIN_SEQ_QUALITY=i
i is the minimum _sequence_ quality within the primer
or oligo (not to be confused with the PRIMER_PAIR_QUALITY
output tag, which is really the value of the objective
function.)
PRIMER_{LEFT,RIGHT}_END_STABILITY=f
f is the delta G of disruption of the five 3' bases of the
primer.
EXAMPLE OUTPUT
--------------
You should run it youself. Use the file 'example' in this
directory as input.
ADVICE FOR PICKING PRIMERS
--------------------------
We suggest referring to: Wojciech Rychlik, "Selection of Primers
for Polymerase Chain Reaction" in BA White, Ed., "Methods in
Molecular Biology, Vol. 15: PCR Protocols: Current Methods and
Applications", 1993, pp 31-40, Humana Press, Totowa NJ
Cautions
--------
Some of the most important issues in primer picking can be
addressed only in Primer3's input. These are sequence quality
(including making sure the sequence is not vector and not
chimeric) and avoiding repetitive elements.
Techniques for avoiding problems include a thorough understanding
of possible vector contaminants and cloning artifacts coupled
with database searches using blast, fasta, or other similarity
searching program to screen for vector contaminants and possible
repeats. Repbase (J. Jurka, A.F.A. Smit, C. Pethiyagoda, and
others, 1995-1996, ftp://ncbi.nlm.nih.gov/repository/repbase)
is an excellent source of repeat sequences and pointers to the
literature. Primer3 now allows you to screen candidate oligos
against a a Mispriming Library (or a Mishyb Library in the case
of internal oligos).
Sequence quality can be controlled by manual trace viewing and
quality clipping or automatic quality clipping programs. Low-
quality bases should be changed to N's or can be made part of
Excluded Regions. The beginning of a sequencing read is often
problematic because of primer peaks, and the end of the read
often contains many low-quality or even meaningless called bases.
When picking primers from single-pass sequence it is often best
to avoid the first 20 base pairs, and to prefer shorter product
sizes or shortened Included Region lengths to avoid low-quality
sequence at the end of the sequence read. In addition, Primer3
takes as input a Sequence Quality list for use with those base
calling programs (e.g. Phred, Bass/Grace, Trout) that provide
this output.
What to do if Primer3 cannot find a primers?
--------------------------------------------
Try relaxing various parameters, including the
self-complementarity parameters and max and min oligo melting
temperatures. For example, for very A-T-rich regions you might
have to increase maximum primer size or decrease minimum melting
temperature. It is usually unwise to reduce the minimum primer
size if your template is complex (e.g. a mammalian genome), since
small primers are more likely to be non-specific. Make sure that
there are adequate stretches of non-Ns in the regions in which
you wish to pick primers. If necessary you can also allow an N
in your primer and use an oligo mixture containing all four bases
at that position.
Try setting the PRIMER_EXPLAIN_FLAG input tag.
DIFFERENCES FROM EARLIER VERSIONS
---------------------------------
Compared to 0.5
---------------
Completely different input format.
It has been reported the 0.5 deleted Ns when they occurred in
primers.
More stringent self-complementarity defaults.
Primer3 selects internal oligos on request (and produces .int
files if requested).
Compared to both 0.5 and v2
---------------------------
The format of the contents of .for, .rev (and .int) files is
different.
Primer3 returns a user-specifiable number of primer pairs (or
triples) sorted by "goodness".
Primer3 will find a primer pair if any acceptable pair exists.
Optional n-based indexing into source sequence.
Use of sequence quality and 3' stability as constraints in primer
picking. Optional positional component to objective function.
Compared to v2
-------------
Tag name changes. However, Primer3 should understand most or
all Primer v2 input tags, and should produce v2-compatible output
tag names when the -v2_compat command-line switch is used.
The one exception is that the PRIMER_RECOMMEND tag is no longer
produced. Instead Primer3 produces the PRIMER_x_EXPLAIN output
tags. The format of the data in this tags is different from the
data in v2's PRIMER_RECOMMEND output tag.
Numerous fixes.
Uses the PRIMER_SELF_ANY and PRIMER_SELF_END parameters to govern
maximum allowable complementarity between left and right primers,
as well as complementarity between copies of a single oligo or
within a single oligo. This behaviour is very close to that of
primer 0.5; self complementarity calculations in v2 were
unreliable.
Primer3 produces much more output information, including the TMs
and self complementarity measures of selected primers.
EXIT STATUS CODES
-----------------
0 on normal operation
-1 under the following conditions:
illegal command-line arguments.
unable to fflush stdout.
unable to open (for writing and creating) a .for, .rev
or .int file (probably due to a protection problem).
-2 on out-of-memory
-3 empty input
-4 error in a "Global" input tag (message in PRIMER_ERROR).
Primer3 calls abort() and dumps core (if possible) if a
programming error is detected by an assertion violation.
SIGINT and SIGTERM are handled essentially as empty input, except
the signal received is returned as the exit status and printed to
stderr.
In all of the error cases above Primer3 prints a message to stderr.
THE PRIMER3 WWW INTERFACE
-------------------------
This distribution contains the Primer3 WWW interface,
primer3.cgi. To execute it you will need perl5 and the perl5
module CGI.pm. Refer to your perl book to locate the perl5
distribution. CGI.pm was written by Lincoln D. Stein and is
available from
http://www.genome.wi.mit.edu/ftp/distribution/software/WWW/ You
will also need to know enough about your operating system and web
server to install a new CGI script, and enough about perl5 to
read the script and figure out how it does what it does. If you
install primer3.cgi at your web site, please change the value of
the $MAINTAINER variable near the top of the file. You might
also have to change the path to the perl5 executable on the first
line. Depending on your primer picking application you might
want to change defaults, and in particular you might want to
specify mispriming libraries in the value of %SEQ_LIBRARY and in
the select boxes named PRIMER_MISPRIMING_LIBRARY and
PRIMER_INTERNAL_OLIGO_MISHYB_LIBRARY. Beyond this you are on
your own.
ACKNOWLEDGMENTS
---------------
The development of Primer3 was funded by Howard Hughes Medical
Institute and by the National Institutes of Health, National
Center for Human Genome Research under grants R01-HG00257 (to
David C. Page) and P50-HG00098 (to Eric S. Lander).
We gratefully acknowledge the support of Digital Equipment
Corporation, which provided the Alphas which were used for most
of the development of Primer3, and of Centerline Software, Inc.,
whose TestCenter memory-error, -leak, and test-coverage checker
helped us discover and correct a number of otherwise latent
errors in Primer3.
Primer3 was written by Helen J. Skaletsky (Howard Hughes Medical
Institute, Whitehead Institute) and Steve Rozen (Whitehead
Institute/MIT Center for Genome Research), based on the design of
earlier versions: Primer 0.5 (Steve Lincoln, Mark Daly, and Eric
S. Lander) and Primer v2 (Richard Resnick). This documentation
was written by Richard Resnick and Steve Rozen. The original web
interface was designed by Richard Resnick.
In addition, following is a partial list of people who kindly
contributed to the design of Primer3
Ernst Molitor
Carl Foeller
The authors of the current version would be pleased to receive
error reports or requests for enhancements. Please send e-mail
to primer3@genome.wi.mit.edu.