Virtual Protein Theme Park
Background of the Invention
Amusement parks are traditionally areas in which a collection of amusement games are placed for peoples entertainment. Typically many of the games have an element of perceived danger and part of the entertainment involves providing sufficient user involvement in the danger that the users adrenalin level is raised.
Where the amusements have some common theme the amusement parks are typically known as "theme parks" with a theme of perhaps, movie background, or of fantasy. Some amusement parks combine several theme parks within the amusement park.
Over time the games, now known as "rides" have increased in technical complexity to provide more perceived danger in the more extreme rides. A century ago the most extreme roller coaster generated heavy positive gravity force (G) and mild negative G on a straight line run with patrons seated in a carriage whereas recent extreme rides generate a little less positive G and a little more negative G on a spiral descending or ascending course with the standing patrons. The design of such a ride is a highly specialized engineering exercise in maintaining the users within a very limited G envelope while ensuring that the carriages have sufficient kinetic energy to run from the top of the ride to the bottom at a safe speed whether loaded or unloaded. While retarders or linear induction motors may assist this process the design is nevertheless an expert field even with computer aided design.
Advances in the use of virtual reality have now made it possible to provide an environment which can imitate a theme park ride with no danger at all, and such rides have been in use for some time. These rides typically imitate a vehicle in which the user is seated, and provide a virtual environment which is that of a space trip, a dangerous flight, a trip back in
time or some other journey with a perceived danger. Other environments such as the CAVE (Cave Advanced Virtual Environment) allow a user to walk through a virtual environment which may include such things as a walk across Moscow's Red Square or through a giant protein.
While such environments are available they do not generally provide an experience which similar to that which is expected by users used to current theme parks with roller coasters, water chutes, railways etc.
Computer games are available which emulate theme parks and which may include roller coasters, chutes and vehicle rides (eg Roller Coaster Tycoon) but these games use a flat screen display and are not virtual reality enabled. Providing the required functionality for a virtual reality environment exceeds the capabilities of these games.
Summary of the Invention
The present invention relates to providing an architecture to build a virtual theme park with the protein structure in which users (students, visitors and game players) can gain various knowledge of bio-molecules through an edutainment ride.
It is known to create three dimensional protein models and programs for doing this are described in "Ribbon: a stereo cartoon drawing program for proteins. J Appl. Cryst. 21 (1988) 572-576" and known from various software programs such as MOLMOL, VMD, Kinemages and CHIME.
In accordance with one object of the invention a method of presenting an emulation of a protein to a viewer comprises presenting at least one three dimensional emulation of the protein as an attraction in a virtual reality theme park wherein the viewer may interact with the emulated protein in real time.
Preferably the protein structure as displayed is derived from molecular information obtained by analysis of a substance and held in a database.
Preferably when the protein as displayed has a main track, path, or backbone the emulation provides a ribbon, tube or half-pipe as the emulation of the track, path or backbone.
Preferably the emulation includes scenery provided by models representing amino acids and H-bonds of the protein.
Preferably the viewer's viewpoint may be associated with the emulated main track and constrained to the main track along it.
Preferably the viewer's viewpoint may be located in or on a virtual transport such as a vehicle or carrier.
Preferably the protein represented is a pathogen such as viruses, prions and bacteria and the viewer's viewpoint is as a host (human or animal) whose species is a vector of or is affected by the pathogen.
Preferably the view is created from databases of molecular structure, of virtual reality representations of hosts and of virtual reality representations of vehicles or carriers.
In another aspect the invention comprises an assemblage of three- dimensional emulations of proteins presented in a virtual reality environment such that the proteins act as attractions in a virtual reality theme park.
Preferably the viewer may transfer between attractions in a virtual reality transport including vehicles and carriers.
Preferably the transport may be a wheeled vehicle, a sliding vehicle or an airborne or spaceborne carrier.
Preferably the transport may be rid by an emulation of a host.
Preferably the vehicle or carrier may be an emulation of an animal, bird or insect. Preferably many users may concurrently use the same virtual reality theme park.
Preferably users may interact with the theme park by use of whole body virtual reality components.
In yet another aspect the invention lies in a theme park wherein at least one of the attractions is based on the appearance of at least part of a protein in a virtual reality environment.
These and various other features as well as advantages which characterize the present invention will be apparent upon reading of the following detailed description and review of the associated drawings.
Brief Description of the Drawings
Fig. 1 shows a block diagram of the manner in which the virtual theme park is generated.
Fig. 2 is a diagram showing the modelling of protein track.
Fig. 3 shows determination protein orientation using the crossing number
Fig.4 shows roller coastering on the half pipe of a protein track.
Fig. 5 shows bicycling inside the full pipe of the protein track.
Fig. 6 shows motorcycling on the half pipe of the protein track.
Fig. 7 shows car racing on the half pipe of the protein track.
Fig. 8 shows riding on a scooter on a protein track
Fig. 9 shows riding on a skateboard on a protein track
Fig. 10 shows skiing on a protein track
Fig. 11 shows aerobatic flying inside the protein surface
Fig. 12 shows mosquito manoeuvring inside the protein surface
Detailed Description
Referring now to Fig. 1 a system for producing virtual protein theme park based on various bio-molecules stored in protein database 101 is shown where the system provides a transport database 102 appearing in the virtual theme park, together with a rider database 103. The protein
database holds protein structure data obtained through either X-ray crystallography or Nuclear Magnetic Resonance (NMR) such data may be held in a known manner, as for instance the format used in the Protein Data Bank (PDB http://www.pdb.org). Data examples in protein database are viruses, bacteria and prions - all organisms which infect animals or humans. The rider database collects hosts (confirmed or suspected) relevant to a protein including pathogens whose structure information is saved in database 101. For instance, chimpanzees are members of the rider database as they are hosts of the HIV protein which is related to AIDS disease. The transport database collects avatars of different types of transportation tools suitable for theme park riding.
The 3D modeller 104 of the virtual protein theme park creates (1) 3D protein models based on the protein database information; (2) 3D transport models based on the transport database information; and (3) 3D rider models based on the rider database.
The models so created have a road track component 105 which can be modelled based on the protein secondary structures and which represent the track surface; a road side scenery component 106 which can be modelled based on the protein amino acids and which represents the scenery local to the track; and other components 107 which include hydrogen bonds (H-bonds), etc.
Referring to Fig. 2 a method for modelling protein tracks and controlling motion when riding along the protein tracks in virtual protein theme park is shown. The method utilizes the polypeptide chain 200 which provides geometric and biological information for modelling of track or backbone trajectory 201, by moving frame 202 along the backbone trajectory and sweeping the path of pipe 203 (half or full) for the protein secondary structure. There are about 20 different types of amino acids and all of them consist of a common central carbon (Cα), a hydrogen atom, an amino group, and a carboxyl group and a varied side chain (R) attached to Cα. Amino acids are joined end-to-end via peptide bonds formed in synthesis, which lead the amino group of one amino acid attached to the carboxyl
group and eliminate water. The main chain of protein consists of Cα, C, and N.
With a given set of points, conventional free-form curve modelling techniques such as B-spline or NURB curve interpolation or approximation can easily generate a smooth curve. From the atom coordinates, the backbone trajectory can be constructed to pass through or near all the backbone atoms. In our approach, however, we just choose the Cα atoms as the sampling points. This is due to the following observations. First, Cα atoms seem more important because they are the connecting points between peptide planes. Second, as the backbone atoms (Cα, C, N, Cα) in a peptide plane can be rather close to each other, the curve interpolating all the atoms may produce zig-zag shapes. Cα atom coordinates are able to produce a reasonable backbone curve while the additional degrees of freedom could be used to control over the normal vectors. Third, reducing sampling points can speed up the curve modelling procedure.
Surface modelling of the backbone is required to present more significant information of the secondary structure. A sweeping technique is used to model the typical helical, strand and loop structures. Thus the moving frames along the trajectory of a protein are expected.
There are many moving frames along a trajectory and many methods have been proposed to compute the frame, for instance those of F. Klok , "Two moving coordinate frames for sweeping along a 3D trajectory." Computer Aided Geometric Design 3, 1986, 217-229; H.W. Guggenheimer , "Computing frames along a trajectory." Computer Aided Geometric Design 6, 1989, 77-78 and B. Juttler , "Generating rational frames of space curves via Hermite interpolation with Pythagorean hodograph cubic splines. " Geometric Modeling and Processing '98, Bookplus Press, 1998, 83-106.
A typical well-known frame is Frenet frame. Traditional differential geometry, as for instance described by G Farin, "Curves and surfaces for computer aided geometric design: a practical guide", 3rd edition, (Academic press), 1993 or I.D. Faux and MJ Pratt, "Computational geometry for design
and manufacturing", (Ellis Horwood Limited, Chichester, UK), 1979, gives a simple way to compute its triplet (tangent, normal and binormal). The Frenet frame, however, has an intrinsic problem related to the normal or binormal vectors. Sometimes these vectors suddenly change directions at some points along the path. This is because the Frenet frame is pathological (see M.G. Wagner and B. Ravani , "Curves with rational Frenet-Serret motion." Computer Aided Geometric Design 15,1997, 79- 101 [3]:) when the curve is perfectly straight for some distance or when the curvature vanishes momentarily, the Frenet frame is undefined. To overcome this a modified approach is taken with the biological properties of the protein backbone to provide a more natural solution. Specifically, the Cα-O vectors are used to provide an initial normal. But the Cα-O vectors may flip in the neighbouring peptide planes. This could lead to a sudden twisting of the sweeping surface. Therefore a simply averaging technique is employed to smooth the normal vector distribution, which allows a natural transition.
For each point on the trajectory, we have tangent, normal and binormal that are unit vectors. Thus we can define tangent, normal and binormal curves. First we generate a NURB curve for the path from the coordination of Cα of a given backbone of a protein. The tangent curve is then created from the path with the shift of the tangent vector. The normal and binormal curves can be constructed in a similar way once the normal and binormal vectors are found. We use the amide plane and the Cα-O vectors to determine the normals. The smooth normals along the path are constructed through a weighted interpolation technique. At the end, the bi-normals along the trajectory are formed by the cross product of tangent and normal vectors.
Fig. 3 shows a method for determining an optimal orientation with a given protein polypeptide. The visualization of the protein is subject to the coordinates of the peptide atoms obtained through X-ray crystallography or NMR. However, the original view of a given protein may not be good for building the protein theme park. Any protein can be rotated in the three
dimensional space. To quantify the orientation, it is possible to introduce a "crossing number" concept based on the projection theory.
Geometrically, protein polypeptides can be represented by polygonal lines (polylines) in 3D space. For a given direction in the 3D space, the 3D polylines can be projected along this direction to form 2D polylines in any planes perpendicular to this direction. Thus, the crossing number, or intersection number, of the 2D projected polylines can be used to characterize the visual complexity to an onlooker of the original 3D polylines. The crossing number of a protein is calculated using an algorithm for a given orientation of a protein from a parametric representation of the polylines. Thus, protein orientation determination is therefore a problem of optimising the crossing numbers in the whole 3D space. The orientation corresponding to the maximum crossing number gives the most complicated view while the minimum crossing number gives the least complicated view.
Using this technique, we can build a virtual protein theme park by determining the protein crossing numbers. For instance, a front view with a minimum crossing number may give the viewers an overall idea about the virtual protein theme park as there are less overlaps of the protein secondary structure in this direction. Naturally, as a viewer moves around the theme park, the number of crossing lines, or the apparent complexity, of the viewed protein will change.
The road track component can be modelled either as half or full pipes in the form of α-helix, β-strand, and loops using a protein structure modelling method based on traversing the protein and determining the protein polypeptide information along the protein. The roadside component can be modelled either as trees or balloons based on the amino acid information. All other components such as H-bonds related to protein secondary structure may be added to the main track.
The transport models are created using CAD techniques and saved in the VRML (virtual reality modelling language) format. Typical transport models include aeroplanes, rockets, racing cars, roller coaster cars, trains, motorcars, motorbikes, bikes, skateboards, skiing boards and skateboard scooters.
The rider models used are mostly determined from the protein given. Should the protein be a pathogen, the rider could be the host (confirmed or suspected) of the pathogen or the relevant disease. Examples include kids as riders of the hand-foot-mouth disease protein.
The graphic renderer 108 then renders all the 3D models and updates the graphics display in either stereo or normal fashion during their life cycle. For stereo view, the rendered graphic images can be in various modes of active stereo, passive stereo, auto stereo or anaglyph. The dynamic motion of the rendered models are governed by the motion controller 109 through the user interface 1 0. Conventional control devices such as a steering wheel could be part of the controller. The transport(s) and rider(s) can be conveniently controlled to provide the motions related to acceleration and deceleration, steering and switching, and free driving. The sound track 111 can be added to the motion of transport(s) and rider(s) along the protein track. The sound track may be created based on the protein amino acids and α-helical, β-strand, and loops which can be mapped into music notes with possible a combination of music equipments. Everything assembled together forms a virtual protein theme park which can be interactively visualized by users 1 2 (game players or students).
Fig. 4 shows a portion of a protein backbone with a train of "carriages" acting as a roller coaster train.
Fig 5 shows a series of bicycles with riders inside a protein chain configured in pipe format.
Fig. 6 shows a motorcycle within a half pipe format.
Fig. 7 shows a car driven by a chimpanzee on a protein backbone configured as a cupped ribbon.
Fig. 8 shows a goat kid riding a scooter through a half pipe.
Fig 9 shows a man riding a skateboard on a half pipe track.
Fig 10 shows a man on a skis in a cupped ribbon track.
Fig 11 shows a bird in flight around a protein molecule
Fig 12 shows the path of a mosquito through a protein molecule.
Within a virtual protein theme park, several games are intended to be included and examples of these are, but are not limited to, roller coasters, bicycling, motorcycling, car racing, scooter riding, skating, skiing, etc.
Industrial applicability
The inventive system of protein theme park can be applied to build real and physical theme park. Computer game on protein theme park can also be developed. The invention has a potential for design of educational toys such as bead-coasters and roller-coasters related to protein structure. Educational models of proteins can also be prototyped.