hITeQ: A new workflow-based computing environment for streamlining discovery. Application in materials science
Introduction
The increasing pressure on chemical industries for enhanced operational efficiency is a persuasive driver for innovation, and new technologies receive a stronger attention by companies and academic laboratories due to their potential impact on current research and development (R&D). On the other hand, the absorption of new techniques and methodologies is not straightforward. Thus, the effectiveness of workflow technology to acquire, analyze, store and integrate disparate apparatus, software, and data in materials science is examined. Such a problematic turns into a real challenge when combined with high throughput (HT) experimentation due to (i) the corresponding reliance on informatics, and the overwhelming amount of data, (ii) the intricacy in describing a solid catalyst [1] and the corresponding use of new computational methodologies since previously developed ones within the life sciences are not flawlessly transferable, (iii) the necessity to cope with various data structure and exchange protocols because very few standards [2] are available. For all these reasons, a reliable and exhaustive capture of related data and processes remains extremely tough and requires tailored tools. Together with InforSense©, the Instituto de Tecnologia Quimica (CSIC-UPV) has dedicated important resources inside TopCombi [3] (“Towards Optimized Chemical Processes and New Materials by Combinatorial Science”), a 5 years integrated project of the 6th PCRD in Europe, started in March 2005 with a budget of 23 M€, and composed of 22 partners, for the harmonization and integration of this information. hITeQ, pronounce “high tech”, is the resulting integrative platform created by the Instituto de Tecnologia Quimica (ITQ) which encapsulates all new development done in ITQ for R&D in materials science and other chemistry fields. As an example, this paper presents the implementation of the recent methodology called Adaptable Time Warping (ATW) [4] for the automatic identification of mixture of crystallographic phases from powder X-ray diffraction data, inside the framework of the platform. The methodology is encapsulated into a so-called workflow, and we explore the benefits of such an environment for streamlining discovery research.
Beside the fact that ATW successfully identifies and classifies crystalline phases from powder XRD for the very complicated case of zeolite ITQ-33 high throughput synthesis process, we stress on the numerous difficulties encountered by both academic laboratories and companies when facing the absorption of new techniques. It is shown how an integrative approach provides a real asset in terms of cost, efficiency, and speed due to its environment that supports well-defined and reusable processes, improves knowledge management, and handles properly multi-disciplinary teamwork, and disparate data structures and protocols. ATW is chosen as application because of zeolites are versatile materials for an increasing number of applications [5] including catalysis; and the discovery of new structures or enlarging the synthesis space, and optimization of existing ones require a considerable experimental effort which can be diminished by using high throughput (HT) technology (usually parallelization and miniaturization) [6] as also done for heterogeneous catalysts [7]. The increase of the amount of experiments implies the data to be automatically analyzed not to slow down the whole process (Fig. 1). One of the challenging tasks is the automatic determination of the crystalline phases contained in each new sample from its powder diffraction data (XRD) because an adapted methodology is still lacking, considering that a specific structure can present differences in the XRD diffractogram, for both the intensity of peaks and the 2θ diffraction angles, depending on its level of crystallinity and its chemical composition.
Firstly, a general description of integrative platforms is given. Then, hITeQ functionalities are described and the ATW methodology is detailed through the examination of the supporting workflow. It is shown how versatile such kind of solution is. Finally, the corresponding benefits are clearly demonstrated.
Section snippets
Integrative and workflow-based platforms
Numerous papers denote new problems and new strategies in various aspects of materials science. Obviously, there is no software that handles all these problems, and therefore it is normal that research centres and companies rely on several software systems with overlap functions. The lack of an integrated architecture motivates the drive to search for new solutions. Workflow technology is a mechanism to integrate data, applications and services, enabling scientists to dynamically construct
Conclusion
hITeQ is a workflow tool which facilitates the design and implementation of workflows, includes a graphic tool for drawing a workflow using icons that represent steps in a process, and a workflow engine that drives the workflow. In contrast to pure automation, workflows assumes and, in fact, requires manual intervention through a work process, such as “input files”, “initial parameters set up”, “selection among results”, etc. In contrast to workflow tools, application automation tools or
Acknowledgement
EU Commission FP6 (TOPCOMBI Project) is gratefully acknowledged.
References (30)
- et al.
Genome Res.
(2003)et al. et al.Bioinformatics
(2004) - et al.
BMC Bioinform.
(2005) - L. Zhao, T. Park, R. Kalyanam, S. Goasguen, SRB Workshop, vol. 1, February 2006, 611...
- et al.
Rev. Sci. Instrum.
(2005)(b)X’Pert HighScore Plus software from PANalytical, see website http://www.panalytical.com/index.cfm?pid=547 (accessed... et al.J. Appl. Crystallogr.
(1988)et al. et al. et al.Powder Diffr.
(1995)J. Appl. Crystallogr.
(1974)(h)RockJock—Uses Microsoft Excel Macros and the Solver function to perform a whole-pattern modified Rietveldtype...(i)See JADE from... - et al.
QSAR Comb. Sci.
(2003)et al.Angew. Chem. Int. Ed.
(2003)et al. - et al.
QSAR Comb. Sci.
(2005) - et al.
ChemEngComm
(2008) - et al.
Nature
(2003)J. Catal.
(2003) - et al.
Micropor. Mesopor. Mater.
(2005)et al.Stud. Surf. Sci. Catal.
(2004)et al.J. Am. Chem. Soc.
(2006)et al.Chem. Mater.
(2006)et al.Comb. Chem. High Throughput Screen.
(2007)
QSAR Comb. Sci.
Angew. Chem. Int. Ed.
Catal. Today
J. Comb. Chem.
J. Neurosci. Methods
Neurocomputing
Comput. Biol. Chem.
Curr. Pharm. Des.
J. Chem. Inf. Comput. Sci.
Cited by (6)
Quaternary mixture designs applied to the development of multi-element oxygen electrocatalysts based on the Ln<inf>0.58</inf>Sr<inf>0.4</inf>Fe <inf>0.8</inf>Co<inf>0.2</inf>O<inf>3-δ</inf> system (Ln = La <inf>1-x-y-z</inf>Pr<inf>x</inf>Sm<inf>y</inf>Ba<inf>z</inf>): Predictive modeling approaches
2011, Catalysis TodayCitation Excerpt :Graphics were prepared using the software applications Mathcad 2000, Origin 8.1 and TeeChart (Steema Soft.) integrated in the hITeQ platform [36]. The experimental data employed can be therefore divided into three groups: (i) compositional data: La, Pr, Sm and Ba lattice stoichiometry; (ii) XRD measurements; and (iii) electrochemical results.
A study on factors affecting the reproducibility of a chemical tongue analysis responding to amino acids
2013, Combinatorial Chemistry and High Throughput ScreeningIntegration of an informatics system in a high throughput experimentation. Description of a global framework illustrated through several examples
2013, Oil and Gas Science and TechnologySynthesis and characterization of extra-large 18-ring zeolite ITQ-33 using phosphonium as structure-directing agent
2012, Gaodeng Xuexiao Huaxue Xuebao/Chemical Journal of Chinese Universities