
NAME
       melting  -  nearest-neighbor  computation  of nucleic acid
       hybridation

SYNOPSIS
       melting [options]

DESCRIPTION
       Melting computes, for a nucleic acid duplex, the  enthalpy
       and the entropy of the helix-coil transition, and then its
       melting temperature.  Three  types  of  hybridisation  are
       possible: DNA/DNA, DNA/RNA, and RNA/RNA.  The program uses
       the method of nearest-neighbors. The set of  thermodynamic
       parameters  can  be easely changed, for instance following
       an experimental breakthrough. Melting is a free program in
       both  sense  of  the term. It comes with no cost and it is
       open-source. In addition it is coded in ISO C and  can  be
       compiled  on  any  operating system. Some perl scripts are
       provided to show how melting can be used  as  a  block  to
       construct more ambitious programs.

       If you use MELTING, please quote one of:

              Le Novre N (2001) MELTING, a free tool to compute the 
              melting  temperature  of  nucleic acid duplex. 
              Bioinformatics, 17: 1226-1227

			  Dumousseau M, Rodriguez N, Juty N, Le Novre N (2012)
			  MELTING, a flexible platform to predict the melting 
			  temperatures of nucleic acids. 
			  BMC Bioinformatics, 13:101.

OPTIONS
       The  options  are treated sequentially. If there is a con
       flict between the value of two options,  the  latter  nor
       mally erases the former.

       -Afile.nn
              Informs  the  program to use file.nn as an alterna
              tive set  of  nearest-neighbor  parameters,  rather
              than  the  default  for the specified hybridisation
              type. The standard distribution of melting provides
              some  files  ready-to-use:  all97a.nn (Allawi et al
              1997), bre86a.nn (Breslauer et al 1986),  san96a.nn
              (SantaLucia  et al 1996), sug96a.nn (Sugimoto et al
              1996), san04a.nn (santalucia et al 2004) (DNA/DNA), 
	      fre86a.nn  (Freier  et  al  1986),xia98a.nn   (Xia 
	      et   al  1998),  (RNA/RNA),  and sug95a.nn (Sugimoto
	      et al 1995), (DNA/RNA).

              The program will look for the file in  a  directory
              specified  during  the installation. However, if an
              environment variable NN_PATH  is  defined,  melting
              will  search  in  this  one  first. Be careful, the
              option -A changes the default parameter set defined
              by the option -H.

       -Ccomplementary_sequence
              Enters  the  complementary sequence, from 3' to 5'.
              This option is mandatory if  there  are  mismatches
              between  the  two  strands.  If it is not used, the
              program will compute it as the  complement  of  the
              sequence entered with the option -S.

       -Ddnadnade.nn
              Informs  the program to use the file dnadnade.nn to
              compute the contribution of dangling  ends  to  the
              thermodynamic  of  helix-coil  transition. The dan
              gling ends  are  not  taken  into  account  by  the
              approximative mode.

       -Ffactor
              This  is  the  a correction factor used to modulate
              the effect of the  nucleic  acid  concentration  in
              the  computation  of  the  melting temperature. See
              section ALGORITHM for details.

       -Gx.xxe-xx
	      Magnesium  concentration  (No maximum concentration
	      for the moment). The effect  of  ions  on  thermod-
	      ynamic  stability  of nucleic  acid duplexes is co-
	      mplex, and the correcting functions are  at  best
	      rough  approximations.The published  Tm  correction
	      formula for divalent Mg2+ ions of  Owczarzy et al (2008)
	      can take in account the competitive binding of mono-
	      valent and divalent ions on DNA. However this formula
	      is only for DNA duplexes.

       -h     Displays a short help and quit with EXIT_SUCCESS.

       -Hhybridisation_type
              Specifies the hybridisation type. This will set the
              nearest-neighbor  set  to use if no alternative set
              is provided by the option -A (remember the  options
              are  read  sequentially).  Moreover  this parameter
              determines the equation  to  use  if  the  sequence
              length  exceeds  the  limit  of  application of the
              nearest-neighbor approach (arbitrarily  set  up  by
              the author). Possible values are dnadna, dnarna and
              rnadna (synonymous), and rnarna.   For  reasons  of
              compatibility  the  values of the previous versions
              of melting A,B,C,F,R,S,T,U,W  are  still  available
              although  strongly deprecated. Use the option -A to
              require an alternative set of thermodynamic parame
              ters.  IMPORTANT:  If  the duplex is a DNA/RNA het
              eroduplex, the sequence of the DNA strand has to be
              entered with the option -S.

       -Iinput_file
              Provides  the  name of an input file containing the
              parameters of the run. The input has to contain one
              parameter  per  line,  formatted  as in the command
              line. The order is not important, as well as  blank
              lines. example:

              ###beginning###
              -Hdnadna
              -Asug96a.nn
              -SAGCTCGACTC
              -CTCGAGGTGAG
              -N0.2
              -P0.0001
              -v
              -Ksan96a

              ###end###

	-ifile.nn
              Informs  the  program to use file.nn as an alterna
              tive set  of  inosine base pair  parameters, rather
              than  the  default  for the specified hybridisation
              type. The standard distribution of melting provides
              some  files ready-to-use:  san05a.nn (Santalucia et 
	      al 2005) for deoxyinosine in DNA duplexes, bre07a.nn
	      (Brent M Znosko et al 2007)for inosine in RNA duple-
	      xes. Note  that  not all the inosine mismatched wob-
	      ble's pairs have been investigated. Therefore it co-
	      uld be impossible to  compute the Tm of a duplex wi-
	      th inosine pairs. Moreover, those inosine pairs are 
	      not taken into account by the  approximative mode.

       -Ksalt_correction
              Permits to chose another correction for the concen
              tration in sodium. Currently, one can chose between
              wet91a, san96a, san98a.  See section ALGORITHM.

       -kx.xxe-xx
	      Potassium  concentration  (No maximum concentration
	      for the moment). The effect  of  ions  on  thermod-
	      ynamic  stability  of nucleic  acid duplexes is co-
	      mplex, and the correcting functions are  at  best
	      rough  approximations.The published  Tm  correction
	      formula for sodium ions of Owczarzy et al. (2008)is there-
	      fore also applicable to buffers containing Tris or
	      KCl. Monovalent K+, Na+, Tris+ ions  stabilize  DNA
	      duplexes with similar potency, and their effects on
	      duplex stability are additive. However this formula
	      is only for DNA duplexes.
                                                              
       -L     Prints   the   legal  informations  and  quit  with
              EXIT_SUCCESS.

       -Mdnadnamm.nn
              Informs the program to use the file dnadnamm.nn  to
              compute the contribution of mismatches to the ther
              modynamic of helix-coil transition. Note  that  not
              all the mismatched Crick's pairs have been investi
              gated. Therefore it could be impossible to  compute
              the Tm of a mismatched duplex. Moreover, those mis
              matches are not taken into account by the  approxi
              mative mode.

       -Nx.xxe-xx
              Sodium  concentration  (between  0  and  10 M). The
              effect  of  ions  on  thermodynamic  stability   of
              nucleic  acid duplexes is complex, and the correct
              ing functions are  at  best  rough  approximations.
              Moreover,  they  are  generally  reliable  only for
              [Na+] belonging to [0.1,1M]. If there are no other
	      ions in solution, we can use only the sodium corre-
	      ction. In the other case, we use the Owczarzy's al-
	      gorithm.

       -Ooutput_file
              The  output is directed to this file instead of the
              standard output. The name of the file can be  omit
              ted.  An  automatic  name is then generated, of the
              form  meltingYYYYMMMDD_HHhMMm.out  (of  course,  on
              POSIX  compliant systems, you can emulate this with
              the redirection of stdout  to  a  file  constructed
              with the program date).

       -Px.xxe-xx
              Concentration  of the nucleic acid strand in excess
              (between 0 and 0.1 M).

       -p
              Return  the  directory  supposed  to  contain   the 
              sets  of  calorimetric  parameters  and  quit  with 
              EXIT_SUCCESS.  If the  environment variable NN_PATH 
              is  set,  it  is  returned.  Otherwise,  the  value 
              defined  by  default  during   the  compilation  is 
              returned.


       -q     Turn off  the  interactive  correction  of  wrongly
              entered parameter. Useful for run through a server,
              or a batch script. Default is OFF (i.e. interactive
              on).  The  switch works in both sens.  Therefore if
              -q has been set in an input file, another -q on the
              command  line  will switch the quiet mode OFF (same
              thing if two -q are set on the same command  line).

       -Ssequence
              Sequence  of one strand of the nucleic acid duplex,
              entered 5' to 3'. IMPORTANT: If  it  is  a  DNA/RNA
              heteroduplex, the sequence of the DNA strand has to
              be entered. Uridine and thymidine are considered as
              identical. The bases can be upper or lowercase. If 
	      There are inosine base pairs in the sequence, a co-
	      mplementary is mandatored.

       -Txxx
              Size threshold  before  approximative  computation. 
              The nearest-neighbour approach will be used only if 
              the length  of  the sequence  is  inferior  to this 
              threshold.

       -tx.xxe-xx
	      Tris buffer  concentration  (No maximum concentrat-
	      ion for the moment). The effect  of  ions  on  the-
	      rmodynamic  stability  of nucleic  acid duplexes is
	      complex, and the correcting functions are  at  best
	      rough  approximations.The published  Tm  correction
	      formula for sodium ions of Owczarzy et al. (2008)is there-
	      fore also applicable to buffers containing Tris or
	      KCl. Monovalent K+, Na+, Tris+ ions  stabilize  DNA
	      duplexes with similar potency, and their effects on
	      duplex stability are additive. However this formula
	      is only for DNA duplexes. Be careful, the Tris+ ion
	      concentration is about half of the total tris buffer
	      concentration.

       -v     Control the verbose mode, issuing a lot more infor
              mation about the current run (try it once to see if
              you can get something interesting). Default is OFF.
              The switch works in both sens. Therefore if -v  has
              been  set  in an input file, another -v on the com
              mand line will switch the verbose  mode  OFF  (same
              thing  if two -v are set on the same command line).

       -V     Displays the version number and quit with EXIT_SUC
              CESS.

       -x     Force  the  program to compute an approximative tm,
              based on G+C content. This option has  to  be  used
              with  caution.  Note that such a calcul is increas
              ingly incorrect  when  the  length  of  the  duplex
              decreases.  Moreover, it does not take into account
              nucleic acid concentration, which is a strong  mis
              take.


ALGORITHM
       Thermodynamics of helix-coil transition of nucleic acid
       The  nearest-neighbor  approach  is based on the fact that
       the helix-coil transition works as  a  zipper.   After  an
       initial  attachment,  the  hybridisation propagates later
       ally.  Therefore, the  process  depends  on  the  adjacent
       nucleotides  on  each  strand  (the  Crick's  pairs).  Two
       duplexes with the same base  pairs  could  have  different
       stabilities,  and  on the contrary, two duplexes with dif
       ferent sequences but identical sets of Crick's pairs  will
       have  the  same thermodynamics properties (see Sugimoto et
       al. 1994).  This program first computes the  hybridisation
       enthalpy  and  entropy  from  the elementary parameters of
       each Crick's pair.

       DeltaH = deltaH(initiation) + SUM(deltaH(Crick's pair))
       DeltaS = deltaS(initiation) + SUM(deltaS(Crick's pair))

       See Wetmur J.G. (1991)  and  SantaLucia  (1998)  for  deep
       reviews  on the nucleic acid hybridisation and on the dif
       ferent set of nearest-neighbor parameters.


   Effect of mismatches and dangling ends
       The mismatching (inosine  mismatches  included) pairs are 
       also taken into account. However the  thermodynamic  para-
       meters are  still  not available for every possible cases
       (notably when both positions are mismatched). In  such a 
       case, the program, unable to compute any relevant result, 
       will quit with a warning.

       The two first and positions cannot be mismatched. in  such
       a  case,  the  result  is unpredictable, and all cases are
       possible. for instance (see Allawi and SanLucia 1997), the
       duplex

       A          T	  			        
        GTGAGCTCAT 	  	 
        TACTCGAGTG		   
       T          A	  	

       is more stable than

       AGTGAGCTCATT	  	
       TTACTCGAGTGA	
       
       The   dangling   ends,   that  is  the  umatched  terminal
       nucleotides, can be taken into account.


   Example
       DeltaH(
       AGCGATGAA-
       -CGCTGCTTT
       ) = DeltaH(AG/-C)+DeltaH(A-/TT)
       +DeltaH(initG/C)+DeltaH(initA/T)
       +DeltaH(GC/CG)+DeltaH(CG/GC)+2xDeltaH(GA/CT)+DeltaH(AA/TT)
       +Delta(AT/TG mismatch) +DeltaG(TG/GC mismatch)

       (The same computation is performed for DeltaS)


   The melting temperature
       Then  the melting temperature is computed by the following
       formula:

       Tm = DeltaH / (DeltaS + Rx ln ([nucleic acid]/F))
       Tm in K (for [Na+] = 1 M )
            + f([Na+]) - 273.15
       correction for the salt concentration (if there are only Na+ 
       cations in the solution) and to get the  temperature in degree 
       Celsius.(In fact some corrections are directly included in the 
       DeltaS see that of SanLucia 1998)


   Correction for the concentration of nucleic acid
       If the concentration of the two strands are similar,  F is 
       1  in  case  of  self-complementary  oligonucleotides,   4 
       otherwise. If one strand is in excess (for instance in PCR 
       experiment),  F is  2  (Actually the formula would have to 
       use the difference of concentrations rather than the total 
       concentration, but if the excess is sufficient,  the total 
       concentration can  be  assumed  to  be  identical  to  the 
       concentration of the strand in excess).

       Note however that MELTING makes the assumption of no self-
       assembly, i.e.  the computation does not take any entropic  
       term to correct for self-complementarity.


   Correction for the concentration of salt
       If there are only sodium ions in the solution, we can use the following
       corrections:
       The correction can be chosen between wet91a, presented  in
       Wetmur 1991 i.e.
       16.6 x log([Na+] / (1 + 0.7 x [Na+])) + 3.85

       san96a presented in SantaLucia et al. 1996 i.e.
       12.5 x log[Na+]

       and san98a presented in SantaLucia 1998 i.e.  a correction
       of the entropic term without modification of enthalpy
       DeltaS = DeltaS([Na+]=1M) + 0.368 x (N-1) x ln[Na+]

       Where N is the length of the duplex.
       
       
   Correction for the concentration of ions when other monovalent
   ions such as Tris+ and K+ or divalent Mg2+ ions are added 
       If there are only Na+ ions, we can use the correction for 
       the concentration of salt(see above). In the opposite case
       , we will use the magnesium and monovalent ions correction
       from Owczarzy et al (2008). (only for DNA duplexes)
       
       [Mon+] = [Na+] + [K+] + [Tris+] 
       
       Where [Tris+] = [Tris buffer]/2. (in the option -t, it is 
       the Tris buffer concentration which is entered). 
       
       If [Mon+] = 0, the divalent ions are the only ions present
       and the melting temperature is :
       
       1/Tm(Mg2+) = 1/Tm(1M Na+) + a - b x ln([Mg2+]) + Fgc x (c 
       		    + d x ln([Mg2+]) + 1/(2 x (Nbp - 1)) x (- e +
		    f x ln([Mg2+]) + g x ln([Mg2+]) x ln([Mg2+]))
       
       where : a = 3.92/100000, b = 9.11/1000000, c = 6.26/100000
       ,d = 1.42/100000,e = 4.82/10000;f = 5.25/10000, g = 8.31/
       100000.
       Fgc is the fraction of GC base pairs in the sequence and 
       Nbp is the length of the sequence (Number of base pairs).
       
       If [Mon+] > 0, there are several cases because we can have
       a competitive DNA binding between monovalent and divalent 
       cations  :
       
       If the ratio [Mg2+]^(0.5)/[Mon+] is inferior to 0.22, mono-
       valent ion influence is dominant, divalent cations can be 
       disregarded and the melting temperature is :
        
       1/Tm(Mg2+) = 1/Tm(1M Na+) + (4.29 x Fgc - 3.95) x 1/100000
                    x ln([mon+]) + 9.40 x 1/1000000 x ln([Mon+])
		    x ln([Mon+])
       where : Fgc is the fraction of GC base pairs in the sequen-
       ce.
       
       If the ratio [Mg2+]^(0.5)/[Mon+] is included in [0.22, 6[,
       we must take in account both Mg2+ and monovalent cations 
       concentrations. The melting temperature is :
       
       1/Tm(Mg2+) = 1/Tm(1M Na+) + a - b x ln([Mg2+]) + Fgc x (c 
       		    + d x ln([Mg2+]) + 1/(2 x (Nbp - 1)) x (- e +
		    f x ln([Mg2+]) + g x ln([Mg2+]) x ln([Mg2+]))
       
       where : a = 3.92/100000 x (0.843 - 0.352 x [Mon+]0.5 x 
       ln([Mon+])), b = 9.11/1000000, c = 6.26/100000
       ,d = 1.42/100000 x (1.279 - 4.03/1000 x ln([mon+]) - 8.03/1000
       x ln([mon+] x ln([mon+]),e = 4.82/10000;f = 5.25/10000, g =
       8.31/100000 x (0.486 - 0.258 x ln([mon+]) + 5.25/1000 x 
       ln([mon+] x ln([mon+] x ln([mon+]).
       
       Fgc is the fraction of GC base pairs in the sequence and 
       Nbp is the length of the sequence (Number of base pairs).
       
       Finally, if the ratio [Mg2+]^(0.5)/[Mon+] is superior to 6,
       divalent ion influence is dominant, monovalent cations can 
       be disregarded and the melting temperature is :
       
              1/Tm(Mg2+) = 1/Tm(1M Na+) + a - b x ln([Mg2+]) + Fgc x (c 
       		    + d x ln([Mg2+]) + 1/(2 x (Nbp - 1)) x (- e +
		    f x ln([Mg2+]) + g x ln([Mg2+]) x ln([Mg2+]))
       
       where : a = 3.92/100000, b = 9.11/1000000, c = 6.26/100000
       ,d = 1.42/100000,e = 4.82/10000;f = 5.25/10000, g = 8.31/
       100000.
       Fgc is the fraction of GC base pairs in the sequence and 
       Nbp is the length of the sequence (Number of base pairs).

   Long sequences
       It is  important  to  realise  that  the  nearest-neighbor
       approach  has  been established on small oligonucleotides.
       Therefore the use of melting in the non-approximative mode
       is  really  accurate  only  for relatively short sequences
       (Although if the sequences are two short, let's  say  <  6
       bp, the influence of extremities becomes too important and
       the reliability decreases a lot).  For long  sequences  an
       approximative  mode  has  been  designed.   This  mode  is 
       launched  if  the sequence length is higher than the value 
       given by the option -T (the default threshold is 60 bp).

       The  melting temperature is computed by the following for
       mulas:

       DNA/DNA:
       Tm = 81.5+16.6*log10([Na+]/(1+0.7[Na+]))+0.41%GC-500/size

       DNA/RNA:
       Tm = 67+16.6*log10([Na+]/(1.0+0.7[Na+]))+0.8%GC-500/size

       RNA/RNA:
       Tm = 78+16.6*log10([Na+]/(1.0+0.7[Na+]))+0.7%GC-500/size

       This mode is nevertheless strongly disencouraged.


   Miscellaneous comments
       Melting is currently accurate only when the  hybridisation
       is performed at pH 71.

       The  computation is valid only for the hybridisations per
       formed in aqueous medium.  Therefore the use of denaturing
       agents   such  as  formamide  completely  invalidates  the
       results.


REFERENCES
       Allawi H.T., SantaLucia J. (1997). Thermodynamics and NMR of 
       internal G-T mismatches in DNA. Biochemistry  36: 10581-10594   

       Allawi H.T., SantaLucia J. (1998). Nearest Neighbor thermodynamics 
       parameters for internal G.A mismatches in DNA. Biochemistry 37: 
       2170-2179

       Allawi H.T., SantaLucia J. (1998).Thermodynamics of internal C.T 
       mismatches in DNA. Biochemistry 26: 2694-2701.

       Allawi H.T., SantaLucia J. (1998). Nearest Neighbor thermodynamics 
       of internal A.C mismatches in DNA: sequence dependence and pH effects.
       Biochemistry 37: 9435-9444.

       Bommarito S., Peyret N., SantaLucia J. (2000).  Thermodynamic parameters 
       for DNA sequences with dangling ends.Nucleic Acids Res 28: 1929-1934

       Breslauer K.J., Frank R., Blocker H., Marky L.A. (1986). Predicting DNA 
       duplex stability from the base sequence. Proc Natl Acad Sci USA  83: 
       3746-3750   

       Freier S.M., Kierzek R., Jaeger J.A., Sugimoto N., Caruthers M.H., 
       Neilson T., Turner D.H. (1986). Biochemistry 83:9373-9377 
 
       Owczarzy R., Moreira B.G., You Y., Behlke M.B., Walder J.A.(2008) 
       Predicting stability of DNA duplexes in solutions containing Magnesium 
       and Monovalent Cations. Biochemistry 47: 5336-5353.  

       Peyret N., Seneviratne P.A., Allawi H.T., SantaLucia J. (1999). Nearest 
       Neighbor thermodynamics and NMR of DNA sequences with internal A.A, C.C,
       G.G and T.T mismatches. dependence and pH effects. Biochemistry} 38:3468
       -3477

       SantaLucia J. Jr, Allawi H.T., Seneviratne P.A. (1996). Improved nearest-
       neighbor parameters for predicting DNA duplex stability.Biochemistry 35: 
       3555-3562   

       Sugimoto N., Katoh M., Nakano S., Ohmichi T., Sasaki M. (1994). RNA/DNA 
       hybrid duplexes with identical nearest-neighbor base-pairs have identical 
       stability. FEBS Letters 354: 74-78   

       Sugimoto N., Nakano S., Katoh M., Matsumura A., Nakamuta H., Ohmichi T., 
       Yoneyama M., Sasaki M. (1995). Thermodynamic parameters to predict stability 
       of RNA/DNA hybrid duplexes. Biochemistry 34: 11211-11216 
  
       Sugimoto N., Nakano S., Yoneyama M., Honda K. (1996).  Improved thermodynamic 
       parameters and helix initiation factor to predict stability of DNA duplexes. 
       Nuc Acids Res 24: 4501-4505  

       Watkins N.E., Santalucia J. Jr. (2005). Nearest-neighbor thermodynamics of 
       deoxyinosine pairs in DNA duplexes. Nuc Acids Res 33: 6258-6267 

       Wright D.J., Rice J.L., Yanker D.M., Znosko B.M. (2007). Nearest neighbor parameters 
       for inosine-uridine pairs in RNA duplexes. Biochemistry 46: 4625-4634

       Xia T., SantaLucia J., Burkard M.E., Kierzek R., Schroeder S.J., Jiao X., 
       Cox C., Turner D.H. (1998). Thermodynamics parameters expanded nearest-neighbor model for formation of RNA duplexes with 
       Watson-Crick base pairs. Biochemistry  37: 14719-14735   


       For review see:

       SantaLucia  J. (1998) A unified view of polymer, dumbbell,
       and oligonucleotide DNA  nearest-neighbor  thermodynamics.
       Proc Natl Acad Sci USA 95: 1460-1465
       
       SantaLucia  J., Hicks Donald (2004) The Thermodynamics of 
       DNA structural motifs. Annu. Rev. Biophys. Struct. 33: 415
       -440

       Wetmur J.G. (1991) DNA probes: applications of the princi
       ples of nucleic acid hybridization.  Crit Rev Biochem  Mol
       Biol 26: 227-259


FILES
       *.nn   Files  containing  the nearest-neighbor parameters,
              enthalpy and entropy, for each Crick's pair.   They
              have to be placed in a directory defined during the
              compilation or targeted by the environment variable
              NN_PATH.

       tkmelting.pl
              A  Graphical  User  Interface written in Perl/Tk is
              available for those  who  prefer  the  'button  and
              menu' approach.

       *.pl   Scripts  are  available to use MELTING iteratively.
              For instance, the script multi.pl permits  to  pre
              dict  the  Tm  of several duplexes in one shot. The
              script profil.pl allow an  interactive  computation
              along  a sequence, by sliding a window of specified
              width.


SEE ALSO
       New  versions  and  related  material  can  be  found   at
       http://www.ebi.ac.uk/compneur/melting/ and at 
       https://sourceforge.net/projects/melting/.

KNOWN BUGS
       The  infiles have to be ended by a blank line because oth
       erwise the last line is not decoded.

       If an infile is called, containing the address of  another
       input file, it does not care of this latter.  If it is its
       own address, the program quit (is it a bug or a feature?).

       In  interactive mode, a sequence can be entered on several
       lines with a backslash

       AGCGACGAGCTAGCCTA\
       AGGACCTATACGAC

       If by mistake it is entered as

       AGCGACGAGCTAGCCTA\AGGACCTATACGAC

       The backslash will be considered as an illegal  character.
       Here  again,  I do not think it is actually a bug (even if
       it is unlikely, there is  a  small  probability  that  the
       backslash could actually be a mistyped base).


COPYRIGHT
       MELTING is copyright (C) 1997, 2013 by Nicolas Le Novre
       and Marine Dumousseau

       This  program  is  free  software; you can redistribute it
       and/or modify it under the terms of the GNU General Public
       License  as  published  by  the  Free Software Foundation;
       either version 2 of the License, or (at your  option)  any
       later version.



Debian GNU/Linux          2013 May 1                          8


ACKNOWLEDGEMENTS
       Nicolas Joly is an efficient and kind debugger and advisor.  
       Catherine Letondal wrote the HTML interface to melting. 
       Thanks to Nirav Merchant, Taejoon Kwon, Leo Schalkwyk, 
       Mauro Petrillo, Andrew Thompson, Wong Chee Hong, Ivano 
       Zara for their bug fixes and comments. Thanks to Richard 
       Owczarzy for his magnesium correction. Thanks to Charles 
       Plessy for the graphical interface files. Markus Piotrowski 
       updated TkMELTING to cover version 4.3. Finally thanks to 
       the usenet helpers, particularly Olivier Dehon and Nicolas 
       Chuche.

AUTHORS
       Nicolas Le Novre
       Babraham Institute, Babraham Research Campus
       Babraham CB22 3AT Cambridge United-Kingdom.
       n.lenovere@gmail.com
       
       Marine Dumousseau
       EMBL-EBI, Wellcome-Trust Genome Campus
       Hinxton CB10 1SD Cambridge United-Kingdom. 
       marine@ebi.ac.uk  

HISTORY

       See  the  file ChangeLog for the changes of the versions 4
       and more recent.

