BIBECHANA 18 (1) (2021) 140-148 

140 
 

BIBECHANA 
ISSN 2091-0762 (Print), 2382-5340 (Online) 

Journal homepage: http://nepjol.info/index.php/BIBECHANA 

Publisher: Department of Physics, Mahendra Morang A.M. Campus, TU, Biratnagar, Nepal 

 

 
Thermal properties of normal and sickled hemoglobin protein 

Jhulan Powrel1,2, Narayan P. Adhikari2* 
1Department of Physics, Butwal Multiple Campus, Tribhuvan University, Nepal 
2Central Department of Physics, Tribhuvan University, Kirtipur Kathmandu, Nepal 
*Email: narayan.adhikari@cdp.tu.edu.np
 

 

Article Information: 

Received: June 29, 2020 

Accepted: August 2, 2020 

 

Keywords:  

Thermodynamic properties 

Sickle cell 

Molecular dynamics 

Hemoglobin protein 
 

 
 

 

ABSTRACT 

Thermodynamic properties of sickled and normal hemoglobin protein 

are considered within the framework of classical molecular dynamics. 

Here we have studied the specific heat capacity and RMSD (Root 

Mean Square Deviation) of both types of hemoglobin protein. Our 

investigation reveals that the specific heat capacity and RMSD for 

oxygenated hemoglobin protein is higher than those of de-oxygenated 

sickle hemoglobin protein. It is also observed that the specific heat  

capacity and RMSD values of sickle hemoglobin protein decrease with 

rise in temperature. 

DOI: https://doi.org/10.3126/bibechana.v18i1.29750 
This work is licensed under the Creative Commons CC  BY-NC License. https://creativecommons.org/licenses/by-nc/4.0/ 
 

 
1. Introduction 

 
 

Hemoglobin is the blood component 

consisting of heme and globin, where heme is 

the prothestic group containing iron and globin 

is the protein compound. It is responsible for 

the transportation of oxygen throughout the 

human body. Heme is synthesized by 

mitochondrion which is fixed with iron. Heme 

is protected by the surrounded globin protein. 

Each single hemoglobin molecule has two 

globin chains, one globin chain is α and the 

other is β. Two hemoglobin molecules 

combine to produce a functional tetramer. In 

inherited hemoglobin the amino acid sequence 

is changed because of the incorrect DNA code 

(HbS). Such a cell having abnormality is a 

sickle cell. Sickle cell disease is the genetic 

disorder in which sudden change of 

hemoglobin protein occurs.  The glutamic acid 

of both the β chain in the sixth position of 

normal hemoglobin protein is replaced by 

valine in the sickle hemoglobin which is 

shown in Figure 1. As a result RBC changes 

from flexible bi-concave to strong elongated 

shape. Thus by carrying less oxygen, the 

hemoglobin becomes fibrous aggregations 

giving the entire red blood cell a sickle shape. 

Sickled hemoglobin protein (HbS) damages 

the sickle erythrocyte and causes the increase 

http://nepjol.info/index.php/BIBECHANA
mailto:narayan.adhikari@cdp.tu.edu.np
mailto:ari@cdp.tu.edu.np
https://doi.org/10.3126/bibechana.v18i1.29750
https://creativecommons.org/licenses/by-nc/4.0/


Jhulan Powrel, Narayan P. Adhikari/ BIBECHANA 18 (1)(2021) 140-148  

141 
 

in the density, reduction in the deformability 

and increase in the adhesivity which shortens 

the life span [1-3]. The polymerization of 

deoxygenated hemoglobin (HbS) is the 

primary and indispensable event in the 

molecular pathogenesis of sickle cell disease. 

There are many symptoms like the attacks of 

pain, sickle cell anemia, swelling in the hands 

and feet, bacterial infections and stroke. Long-

term pain may develop as people get older 

because of sickle cell anemia. On the other 

hand,  heterozygote for the sickle gene are 

relatively protected against the danger of dying 

by malaria, as now firmly established through 

a number of clinical field studies from 

different parts of Africa and Asia(Terai region 

of Nepal)[4].  
 

Figure 2(a) shows the glutamate 6 of α chain 

in hemoglobin protein which is undergoes a 

mutation to valine. This mutation changes the 

charge on the surface of hemoglobin. The 

mutant protein is called sickle cell hemoglobin 

(HbS). Valine residue 6 of the β chain of 

deoxy-HbS lies on the surface of the protein.  

This hydrophobic residue, present in each of 

the β chain, forms a hydrophobic contact in 

the neighboring β chain of another 

hemoglobin molecule [1, 6]. Figure 2(b) 

shows the mutation that replaces a glutamate 

residue by a valine residue which reduced 

the solubility of the protein. 
 

Luzzatto L. studied the structural abnormality 

of sickle hemoglobin (HbS) and noticed as this 

was the first time that a single amino acid 

replacement in a protein causes a serious 

disease [7]. A group of scientists, Stradner and  

 

 

 

 

 

 

 

 

Fig. 1: Normal (glu) and sickle cell (val) hemoglobin [5]. 

co-worker studied molecular dynamical 

simulations [1] and developed the concept of 

the study from soft matter physics to complex 

protein mixtures [8].A potential antisickling 

drug is still to be discovered [9] as per 

pathophysiological study in the last 100 years 

of sickle cell disease. Another group, Xianqiao 

Wang and Rodney D. developed a virtual 

mechanical model for determining the 

mechanical and thermal behavior of different 

soft matter by using the steps of optimization 

of energy, making of equilibration in the 

system and production run [10]. An electro-

chemistry based study designed a technique to 

control and monitor the polymerization of 

sickle-cell hemoglobin (HbS) by looking the 

change in turbidity during the depletion of 

oxygen in a small volume of custom-built thin-

layer electrochemical cell. The cell allowed the 

investigation of HbS polymerization as a 

function of HbS concentration, temperature 

and solution pH [11]. 

Due to bad impact of sickle cell disease, this 

research work got attention on the study of 

thermal and transport property of sickle 

hemoglobin and normal hemoglobin protein. 

Literature review also reveals that the 

mechanical, thermal and transport properties of 

normal hemoglobin and sickle hemoglobin 

protein around human body temperature (310 

K) are not studied yet. 
 

The paper is organized with methods and 

methodology part in section 2, computational 

simulation part in section 3, results with 

discussion in section 4 and finally the 

conclusions at the last. 

  

 

 

 



Jhulan Powrel, Narayan P. Adhikari/ BIBECHANA 18 (1)(2021) 140-148  

142 
 

 

 
 

Fig. 2: (a) Normal hemoglobin with glutamic acid in 6th position of beta chain (left) and (b) 

Sickle hemoglobin protein with valine in the 6th position of beta chain (right). 

 
2. Methods and Methodology 
 

Theory 
 

 
 

In molecular dynamics (MD), a hemoglobin 

protein molecule is considered inside the 

water for simulation. Protein is the 

combination of a number of atoms which 

behaves as pair-wise interaction in between 

two atoms in the molecule [12]. Each pair of 

atoms is interacted by bonded and non-

bonded potentials. The molecular dynamic 

simulation calculates the motion of the atoms 

in molecules using Newtonian dynamics. In 

standard molecular dynamics simulations the 

forces may come from classical inter atomic 

potentials [13]. The non-bonded van der 

Waals and electrostatic are the interactive 

forces among the particles in the hemoglobin 

molecules. The total potential is obtain as, 

 

Vtotal = [Vbonded]  +  [Vnonbonded]               (1)                                                 

Or, 

Vtotal = [Vbond + Vangle + Vimproper +    

                                  Vproper] + [VLJ + Vcoloumb]             (2) 

 

Now the bond stretching potential is given by, 

 

        (3) 

where 
b

ijK force is constant, ijr is separation 

between the two bounded particles and ijb  is the 

equilibrium bond length. Again the bond angle 

potential is written as, 
 

)4()(
2

1
)( 20

ijkijkijkijangle KV   −=
 

 

where ,ijk 0

ijk and 

ijkK are the bond angle, 

equilibrium bond angle and the force constant 

respectively. Now the improper dihedral potential 

energy is, 
 

)5()(
2

1
)( 2

0  −= ijklijklimproper KV
 

where 
0 is the equilibrium improper  angle and 

K is the constant.  Similarly the dihedral potential 

is,
                

)6())cos(1(
2

1
)( sijklijklproper nKV   −+=  

 

where ijkl
 
is the proper dihedral angle, s is an 

angle at which the potential is maximum, n is the 

multiplicity and  K
 
is the barrier height. 

 

( ) .
2

1
)(

2
ijijij

b
ijbond brKrV −=



Jhulan Powrel, Narayan P. Adhikari/ BIBECHANA 18 (1)(2021) 140-148  

143 
 

Although Lennard-Jones potential(L-J) is 

not suitable for system where strong 

localized bonds may form (as in covalent 

system) or there is a delocalized electron sea 

where the ions remain (as in metals), the L-J 

potential helped us to understand basic points 

in many areas of condensed matter physics. 

The L-J potential for interaction between a 

pair of atom is given by [14], 

)7(4

612






















−








=

rr
VLj


  

where ε is the depth of the attractive well, σ 

refers to the inter-particle distance and 

defines the strength of the interaction where 

the potential changes sign and r defines the 

length scale. The Coulomb electrostatic 

interaction between two point charges qi and 

qj is defined by Coulomb potential as, 

)8(
4 2r

qq
V

m

ji
Coloumb


=  

Root Mean Square Deviation (RMSD): 

RMSD is the level of equilibration and gives 

the deviation of the molecules from the 

defined position in space. It also gives the 

idea of occurrence of the reaction in 

chemical bonding. The larger the deviation 

from mean position less will be the stability 

of the system. RMSD values are calculated 

for all atoms of the hemoglobin protein 

backbone for the entire protein and for the 

protein excluding the last five residues[15] , 

which is the numerical measure of the 

structural difference of two states given by, 

( )
( ) ( )

)9(

2

1

t

jj
N

j
N

trtr

tRMSD







 −

=
= 


 

With 

( ) ( ) )10(
1

1 j
N
j

t
j tr

N
tr t

 ==  

where Nα is the number of atoms whose 

positions are compared, Nt is the number of 

time steps over which atomic positions are 

compared, →−rα(tj) is the position of atom α at 

time tj and ( )jtr is the average value of the 

position of atom to which the positions →−r α(tj) 

are being compared[15]. 

Specific heat Capacity: 

The specific heat capacity is the amount of 

heat required for raising the temperature of 

an object per degree of temperature increase 

per unit mass. Thus in canonical ensemble 

the specific heat capacity is [16], 

)11(
 Tk 2

B

22
BB

v

EE
c

−
=

where 
2

BE  is square of the  average of the  

total energies and 
2

BE   is the  average of  the  

square of the total energy and Bk  the 

Boltzmann constant. 
 

3. Computational details 
 
 

The proposed research is designed for working 

in theoretical computational method using 

molecular dynamics. Hemoglobin protein of 

sickle and normal cell are obtained from the 

RCSB (Research Collaboratory for Structural 

Bioinformatics).org in PDB (Protein Data 

Bank) form and then simulation is completed 

using NAMD (NAno  scale Molecular 

Dynamics) and VMD (Visual Molecular 

Dynamics) software by following the steps of 

structure making, energy minimization and 

making of equilibration in the system. After 

performing various simulation steps the 

output data are used for the calculations of 

RMSD and specific heat capacity of both of 

the hemoglobin protein. For this simulation 

six systems are designed with four molecules 

which are then solvated with water using 

charmm force field (CHARMM 36) under 

following parameters: A system of 

oxygenated hemoglobin protein (a3n) with 

4546(without water) atoms under the 

interaction of the bond potential and non-



Jhulan Powrel, Narayan P. Adhikari/ BIBECHANA 18 (1)(2021) 140-148  

144 
 

bonded potential inside a  box  with  cut  off  

distance of  12.0 Å,  pressure  of  1 bar and 

temperature  310 K. A system of oxygenated 

hemoglobin protein(e83) with 19167(without 

water is 4668) atoms under the interaction of 

the bond potential and non-bonded potential 

inside a box with cut off distance of 12.0 Å, 

pressure of 1 bar and temperature 310 K. A 

system of oxygenated hemoglobin 

protein(uvx) with 1863 atoms under the 

interaction of the bond potential and non-

bonded potential inside a box with cut off 

distance of 12.0 Å, pressure of 1 bar and 

temperature 310 K. Similarly, for other three 

systems of sickle hemoglobin protein(HbS) 

there are 9112 atoms in each and interacting 

at the same potential as before with pressure 

at 1 bar, cut off distance of 12.0 Å and at 

temperatures of 305 K, 310 K and 315 K. 

After equilibration of all the above six 

system they are energetically minimized at 

their corresponding constant temperature 

which provides their RMSD values. At last 

systems are further modified for NVE and 

NVT simulation run in which we use the 

temperature coupling feature of NAMD to 

set the temperature of the molecules in the 

outer layer of the sphere to 300 K, while the 

rest of the bob is set to temperature of 330 K. 
 

4. Result and Discussion 
 

Present work was to study the RMSD as well 

as specific heat capacity of normal and sickle 

hemoglobin protein. The RMSD was 

calculated using the equation (9). 
 

We plotted the RMSD of hemoglobin protein 

against the time steps as follows; where Figure 

3(a) and 3(b) indicates the graph of root mean 

square deviations of oxygenated sickle 

hemoglobin protein and de-oxygenated 

hemoglobin protein at temperature of 310 K 

and obtained the  RMSD as 0.8 Å and 1.25 

Å respectively. They were  found to be 

inc reased  initially with small fluctuation 

and then remain constant for both at various 

values, i.e. the RMSD of deoxygenated 

hemoglobin found to be higher than of 

oxygenated sickle hemoglobin protein. 
 

RMSD plot of other system of hemoglobin 

protein were also plotted where Figure 4(a) 

and 4(b) indicates the root mean square 

deviations of oxygenated hemoglobin protein 

and sickle hemoglobin protein at temperature of 

310 K. For which the RMSD found to be 

increased initially with small fluctuation and 

then remain constant for both at various values 

of 1.3 Å and 0.75 Å respectively, i.e. the 

RMSD of oxygenated hemoglobin is higher 

than of sickle hemoglobin protein. 
 

Similarly, RMSD for sickle hemoglobin 

protein at various temperatures were studied, 

where Figure 5(a) and 5(b) indicating the 

root mean square deviations of sickle 

hemoglobin protein at 305 K and 315 K 

respectively. They are increased initially with 

small fluctuation and then remain constant 

for both at various values of 1.5
 
Å and 1.4 Å 

respectively i.e. the RMSD of sickle 

hemoglobin protein at lower temperature is 

higher than of sickle hemoglobin protein at 

higher temperature. The above Figure 3(a), 

3(b), 4(a) and 5(a), are the RMSD graph at 

310 K for carbon monoxy-hemoglobin(s), de-

oxygenated hemoglobin protein, truncated 

heme-ligand hemoglobin and sickle hemoglobin 

protein respectively. Correspondingly their 

RMSD were estimated as 0.8
 
Å, 1.2

 
Å, 1.3 Å 

and 0.75
 

Å, i.e.   the  RMSD  of  sickle 

hemoglobin  is  least(0.75
 
Å) and of truncated  

ligand  hemoglobin  is  maximum(1.3
 

Å). 

Thus, deoxygenated sickle hemoglobin with 

fibrous nature shows least deviation from the 

mean position, whereas of truncated ligand 

hemoglobin with large no of hydrogen bond 

and hydrophobic potential move faster with 

more RMSD value. Similarly normal 

oxygenated and deoxygenated hemoglobin 



Jhulan Powrel, Narayan P. Adhikari/ BIBECHANA 18 (1)(2021) 140-148  

145 
 

3 

have their RMSD 1.3 Å and 1.25 Å respectively 

which are higher than both the oxygenated (0.8 Å) 

and deoxygenated sickle hemoglobin (0.75 Å).  

Again at   temperatures of 305 K, 310 K and 315 K 

(Figure 4(b), 5(a) and Figure.5 (b)) of de-

oxygenated sickle hemoglobin (HbS) [17] the 

RMSD are 1.5 Å, 0.75 Å  and 1.4 Å respectively.   
 

Figure 6(a) and 6(b) are energy fluctuation 

and energy distribution graph during NVE 

and NVT run. They gives the energy 

deviation from the mean [Fig 6(a)] and 

energy distribution pattern [Fig 6(b)] i.e. 

more the energy more is the velocity as well 

as temperature of the system of particles and 

hence system of protein behaving as 

thermometer. Computationally by using 

NAMD code the average of square of the 

total energies and square of the average total 

energy of de-oxygenated hemoglobin 

protein(e83) were obtain as 6722047.08 

kcalmol−1  and  6714152.23 kcalmol−1  

 

 

 

respectively. On using these values the 

specific heat capacity of de-oxygenated 

hemoglobin protein was estimated as 2903 

Jkg−1C−1 (Table 1). Now for the calculation 

of specific heat capacity (Cv), we have used 

the following theoretical and simulation 

parameters under the assumption of classical 

molecular dynamics: kB=0.00198657 kcal /mol-

K, σ ∼ 8 K, T=317 K, J=1.43846 ×1020 

kcal/mol. 
 

The specific heat capacity of normal and 

sickle hemoglobin protein were calculated by 

using the equation (11) and are tabulated in 

Table 1 and Table 2 for all the six systems. 

The average specific heat capacity of sickle 

hemoglobin 67.1906
3

122719662527
( =

++
 J/kg-C 

[Table 1]) was less than of normal 

hemoglobin's (3587 J/kg-C, 2903 J/kg-C 

and 3204 J/kg-C) [Table:2]. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 3: (a) RMSD graph of oxygenated sickle hemoglobin protein (left). (b) RMSD graph of 

deoxygenated hemoglobin protein (right). 

 

 

  

 

 

 



Jhulan Powrel, Narayan P. Adhikari/ BIBECHANA 18 (1)(2021) 140-148  

146 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 4: (a) RMSD graph of oxygenated hemoglobin protein (left). (b) RMSD graph of sickle 

hemoglobin protein (right). 

 

 

 

 

 

 

 

 

Fig. 5: (a) RMSD graph of sickle hemoglobin protein at 305 K (left). (b) RMSD graph of 

sickle hemoglobin protein at 315 K (right). 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 6: (a) Energy flocculation of de-oxygenated hemoglobin protein (b) Maxwell energy 

distribution for de-oxygenated hemoglobin protein. 

 



Jhulan Powrel, Narayan P. Adhikari/ BIBECHANA 18 (1)(2021) 140-148  

147 
 

Table 1: For normal and sickle cell hemoglobin protein at 310 K of temperature. 

For RMSD(Å) Cv(Jkg−1C−1) 

Oxygenated sickle hemoglobin protein) 0.8 3587 

De-oxygenated hemoglobin protein 1.25 2903 

Oxygenated hemoglobin protein 1.3 3204 

Sickle hemoglobin protein(2hbs) 0.75 1966 

 

Table 2: For sickle cell hemoglobin protein (2hbs) at different temperature. 

For Temp.(K) RMSD(Å) Cv(Jkg−1C−1) 

Sickle hemoglobin protein(2hbs) 305 1.5 2527 

Sickle hemoglobin protein(2hbs) 310 0.75 1966 

Sickle hemoglobin protein(2hbs) 315 1.4 1227 

 

 

5. Conclusion and Concluding Remarks 
 

The  RMSD  of  normal  and  sickle  

hemoglobin  protein was calculated  within  

0.75 Å to  1.5 Å which  indicates  that  the  

proposed  system  is  in the  valid  range(<  4 Å)  

for  molecular dynamical study. This study 

also reveals that the HbS at 305 K is most 

unstable (RMSD=1.5 Å) among the considered 

system. The RMSD and specific heat capacity 

(Cv) of de-oxygenated hemoglobin protein at 

310 K are estimated as 1.25 Å and 2903 

J/kg-C [Table 1] respectively. It also obvious that 

the de-oxygenated hemoglobin protein 

(average value=1906.67J/kg-C) has lower 

specific heat capacity than oxygenated (first 

three Cv of Table 1). It shows that at higher 

(315 K) temperature above 310 K the RMSD 

of sickle hemoglobin protein is increased due 

to less polymerization (sickling) of protein 

but increased in RMSD at lower temperature 

(305 K) is still not understood. 

We found essentiality of additional research 

work on the free energy calculation along 

with the mechanical and transport properties of 

sickle hemoglobin protein   for its depth 

knowledge. 
 

Acknowledgments 
 

JP acknowledges UGC, Nepal for providing 

financial support (Grant No. S&T-13-75/76) 

and Butwal Multiple Campus, Butwal 

(Tribhuvan University) for providing Ph. D. 

leave. This work is also under the grant of 

UGC, Nepal for which NPA acknowledges the 

UGC Award no. CRG-73/74-S&T-01. 
 

References 
 

[1]  A.  Stradner  et  al,  New  insight  into  

cataract  formation  â  enhanced  stability 

through mutual attraction, Physics 

Department and Fribourg Center for 

Nanomaterials, University of Fribourg, CH-

1700 Fribourg, Switzerland (2007). 



Jhulan Powrel, Narayan P. Adhikari/ BIBECHANA 18 (1)(2021) 140-148  

148 
 

[2] W. A. Eaton and J. Hofrichter, Sickle cell 

hemoglobin polymerization, Advances in 

Protein Chemistry  40 (1990) 63-279.  

http://doi.org/10.1016/s0065-3233(08)60287-9. 
 [3] F. Ferrone, R. L. Nagel, Sickle hemoglobin 

polymerization. In: Disorders of 

hemoglobin: Genetics, pathophysiology, 

clinical management, Steinberg  MH, For- 

get BG, Higgs D, Nagel RL (Eds), 

Cambridge University Press (2000). 

[4] Luzzatto, Lucio., Sickle cell anaemia and 

malaria, Mediterranean Journal of 

Hematology and Infectious Diseases (4)1 

(2012) e2012065. 

http://doi.org/10.4084/MJHID.2012.065. 

 [5] D. C. Rees, T. N. Williams, M. T. Gladwin, 

Sickle-cell disease, The Lancet              

376(9757) (2010) 2018-2031. 

[6] J. Robert, J. Ouellette, David Rawn, Organic 

Chemistry (Second Edition) Structure, 

Mechanism, Synthesis (2018) pp. 929-971. 

[7] W. D. Callister, Materials Science and 

Engeenering, An Introduction, John Wiley 

Sons.Inc. US, (2007). 

[8] Retrieved Sept. 22th, 2018 from 

www.google.com/Sickle Cell Anemia. 

[9] Mahran et al., Beni-Suef University Journal 

of Basic and Applied Sciences, Springer 

open, Egypt (2019). 

  https : //doi.org/10.1186/s43088 − 019 − 0016 − x 

[10] Wang Xianqiao and D. Rodne, Avereabtt 

Molecular dynamics simulations indicate 

that deoxyhemoglobin,  oxyhemoglobin, 

carboxy-hemoglobin, and  glycated haemoglobin 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

hemoglobin under compression and shear 

exhibit an anisotropic mechanical behavior, 

Sumith Yesudasan ORCID Icon 

(2017)1417-1429. 

https://doi.org/10.1080/07391102.2017.1323674. 

[11] Iqbal, Zeshan et al., Investigation of sickle-cell 

polymerisation under electrochemical control, 

Chemphyschem: A European Journal of  

Chemical Physics and Physical 

Chemistry 14 (10)  (2013)  2143-8 .  

https://doi.org/10.1002/cphc.201300203. 

[12] G. R. Serjeant, Sickle-cell disease. The Lancet 

350(9079) (1997) 725-730 

[13] M. P. Allen and D. J. Tildesley., Computer 

Simulation of Liquids, Oxford University 

Press, New York (1987). 

[14] D. Frenkel, B. Smit, Understanding 

Molecular Simulation From Algorithms to 

Applications 2nd ed; Academic Press; San 

Diago, USA (2002). 

[15] C. James Phillips, Rosemary Braun, Wei 

Wang, James Gumbart, Emad Tajkhorshid, 

Elizabeth Villa, Christophe Chipot, Robert 

D. Skeel, Laxmikant Kale, and Klaus 

Schulten. Scalable molecular dynamics with 

NAMD. Journal of Computational 

Chemistry 26 (2005)1781-1802. 

www.ks.uiuc.edu/Training/Tutorials(RetrivedonJu

ne10th2020 

[16] K. Huang, Statistical Mechanics, John Wiley 

& Sons, New York (1987).  

[17] J. C. Phillips et al., Scalable molecular dynamics 

with NAMD, J. Comp. Chem. 26 (2005) 1781-

1802. 

 

 

http://www.google.com/Sickle
http://www.ks.uiuc.edu/T
http://www.ks.uiuc.edu/T