Computational Design Framework Research Articles

A computational framework to empower probabilistic protein design.

Motivation: The task of engineering a protein to perform a target biological function is known as protein design. A commonly used paradigm casts this functional design problem as a structural one, assuming a fixed backbone. In probabilistic protein design, positional amino acid probabilities are used to create a random library of sequences to be simultaneously screened for biological activity. Clearly, certain choices of probability distributions will be more successful in yielding functional sequences. However, since the number of sequences is exponential in protein length, computational optimization of the distribution is difficult.Results: In this paper, we develop a computational framework for probabilistic protein design following the structural paradigm. We formulate the distribution of sequences for a structure using the Boltzmann distribution over their free energies. The corresponding probabilistic graphical model is constructed, and we apply belief propagation (BP) to calculate marginal amino acid probabilities. We test this method on a large structural dataset and demonstrate the superiority of BP over previous methods. Nevertheless, since the results obtained by BP are far from optimal, we thoroughly assess the paradigm using high-quality experimental data. We demonstrate that, for small scale sub-problems, BP attains identical results to those produced by exact inference on the paradigmatic model. However, quantitative analysis shows that the distributions predicted significantly differ from the experimental data. These findings, along with the excellent performance we observed using BP on the smaller problems, suggest potential shortcomings of the paradigm. We conclude with a discussion of how it may be improved in the future.Contact: fromer@cs.huji.ac.il

Using an Adjoint Approach to Eliminate Mesh Sensitivities in Computational Design

An algorithm for efficiently incorporating the effects of mesh sensitivities in a computational design framework is introduced. The method is based on an adjoint approach and eliminates the need for explicit linearizations of the mesh movement scheme with respect to the geometric parameterization variables, an expense that has hindered practical large-scale design optimization using discrete adjoint methods. The effects of the mesh sensitivities can be accounted for through the solution of an adjoint problem equivalent in cost to a single mesh movement computation, followed by an explicit matrix-vector product scaling with the number of design variables and the resolution of the parameterized surface grid. The accuracy of the implementation is established and dramatic computational savings obtained using the new approach are demonstrated using several test cases. Sample design optimizations are also shown.

Diffusion of Creative Design: Gatekeeping Effects

A computational framework for design is presented to show that certain social structures can determine how novel solutions are created and spread. This paper suggests that creativity transcends the individual inasmuch as situational factors such as the role of gatekeepers can determine who is considered creative in a society.

A hybrid cross-mapping neural network model for computational intelligent design

In order to assist the designer in the design process, it is necessary to develop a new intelligent computational methodology that involves the integration of design, analysis and evaluation, and optimisation. In this paper, a hybrid cross-mapping neural network model (HCMNN) is proposed to support design, analysis and evaluation, and optimisation in the design process, which integrates a back-propagation neural network (BPNN) and a Hopfield neural network (HNN). The BPNN is used for representing design patterns, training classification boundaries, and outputting the net weight values to the HNN, and then the HNN uses the calculated weight values to evaluate and modify or redesign the design patterns. The developed system provides a unified computational intelligent design framework. The system has self-modifying or self-learning functions. Within the system, only one network is needed to train for accomplishing design analysis and evaluation, rectification/modification and optimisation tasks in the design process. Finally, two case studies are provided to verify the proposed model.

A neuro-fuzzy hybrid scheme for design and simulation of human machine systems

In this article, a new neuro-fuzzy hybrid approach to human workplace design and simulation is proposed. Problems related to human workplace design such as human-machine modeling, measurement and analysis, workplace layout design and planning, workplace evaluation and simulation are discussed in detail. The complex human-machine interactions in workplace design are described with human and workstation parameters within a comprehensive human-machine system model. Based on this model, procedures and algorithms for workplace design, ergonomic evaluation, and optimization are presented in an integrated framework. With a combination of individual neural and fuzzy techniques, the neuro-fuzzy hybrid scheme implements fuzzy if-then rules block for workplace design and evaluation by trainable neural network architectures. For training and test purposes, simulated assembly tasks are carried out on a self-built multiadjustable laboratory workstation with a flexible PEAK Motus motion measurement and analysis system. The trained fuzzy neural networks are capable of predicting the operator's posture and joint angles of motion associated with a range of workstation configurations. They can also be used for design/layout and adjustment of manual assembly workstations. The developed system provides a unified, intelligent computational framework for human-machine system design and simulation. In the end, case studies for workplace design and simulation are presented to validate and illustrate the developed neuro-fuzzy design scheme and system.

Formulating and solving parametric design problems involving non-interference constraints

Improvements in computer-aided design (CAD) tools can significantly increase designer productivity, since the ability to explore a variety of possible designs quickly and effectively is essential for a designer. Using an optimization tool, systematic exploration of design spaces can be achieved readily. The general goal of the work presented here is to aid design by combining the strengths of optimization techniques with those of CAD systems. The specific objective of this paper is to introduce goal directed geometry (GDG) as a computational framework for parametric design, aiding the formulation of engineering problems with geometric considerations and their solution with a multi-objective optimization package. Using GDG, ‘What if’ questions can be posed and answered in a systematic fashion. Specific issues to be addressed include the development of a general parametric design problem formulation, development of static and dynamic geometric non-interference constraints for use in this formulation, and investigation of the efficacy of the adaptive linear programming (ALP) multiobjective optimization algorithm in solving such problems. Two examples are presented, one each to illustrate the use of the static and dynamic non-interference constraints. Results demonstrate that the GDG formulation can be applied readily to a wide variety of parametric design problems. Additionally, the ALP algorithm successfully navigates around geometric constraints, although care must be taken when linearizing highly non-linear design spaces.

A methodology and computational framework for concurrent product and process design

We describe a methodology for Concurrent Product and Process Design and a computational framework that supports it. The methodology is based on the premise that manufacturability is best assured by simultaneously designing a part or assembly and the process used to fabricate it. The designer works in “manufacturing modes”, adding features to the design or manipulating the process plan itself. The computational framework to support this methodology is inspired by human team design and consists of cooperating experts (human and computational) that critique and refine a design from multiple perspectives (e.g. tooling, fixturing, path planning, materials).