A neural network based tool for semi-automatic code transformation

A neural network based tool has been developed to assist in the process of code transformation. The tool offers advice on appropriate transformations within a knowledge-driven, semi-automatic parallelisation environment. We have identified the essential characteristics of codes relevant to loop transformations. A Kohonen network is used to discover structure in the characterised codes thus revealing new knowledge that may be brought to bear on the mapping between codes and transformations or transformation sequences. A transform selector based on this process has been developed and successfully applied to the parallelisation of sequential codes.

Reuse of construction waste: performance under repeated loading

The use of recycled aggregates has increased greatly over the last decade owing to enhanced environmental sensitivities. The level of performance required by such materials is dependent upon the applications for which they are used. Many recycled construction wastes have adequate shear strength in relation to various geotechnical applications. However, a possible drawback of these materials is the risk of crushing during repeated loading. The work reported in this paper examined two waste materials: crushed concrete and building debris, both regarded as construction wastes. Tests were also performed on traditionally used crushed rock, in this case basalt. The materials were subjected to repeated loading in a large direct shear apparatus. The amount of crushing was quantified by performing particle size analysis of the tested material. The results have shown that both recycled construction wastes were susceptible to particle crushing. The amount of crushing was influenced by both the vertical pressure and the number of loading cycles. This leads to a marked decrease in peak friction angle

Continuous MLL-ENL expression is necessary to establish a "Hox Code" and maintain immortalization of hematopoietic progenitor cells

The t[(11;19)(p22;q23)] translocation, which gives rise to the MLL-ENL fusion protein, is commonly found in infant acute leukemias of both the myeloid and lymphoid lineage. To investigate the molecular mechanism of immortalization by MLL-ENL we established a Tet-regulatable system of MLL-ENL expression in primary hematopoietic progenitor cells. Immortalized myeloid cell lines were generated, which are dependent on continued MLL-ENL expression for their survival and proliferation. These cells either terminally differentiate or die when MLL-ENL expression is turned off with doxycycline. The expression profile of all 39 murine Hox genes was analyzed in these cells by real-time quantitative PCR. This analysis showed that loss of MLL-ENL was accompanied by a reduction in the expression of multiple Hoxa genes. By comparing these changes with Hox gene expression in cells induced to differentiate with granulocyte colony-stimulating factor, we show for the first time that reduced Hox gene expression is specific to loss of MLL-ENL and is not a consequence of differentiation. Our data also suggest that the Hox cofactor Meis-2 can substitute for Meis-1 function. Thus, MLL-ENL is required to initiate and maintain immortalization of myeloid progenitors and may contribute to leukemogenesis by aberrantly sustaining the expression of a "Hox code" consisting of Hoxa4 to Hoxa11.

A meta-heuristic approach to parallel code generation

The efficient generation of parallel code for multi-processor environments, is a large and complicated issue. Attempts to address this problem have always resulted in significant input from users. Because of constraints on user knowledge and time, the automation of the process is a promising and practically important research area. In recent years heuristic approaches have been used to capture available knowledge and make it available for the parallelisation process. Here, the introduction of a novel approach of neural network techniques is combined with an expert system technique to enhance the availability of knowledge to aid in the automatic generation of parallel code.

Ionic Liquid-Based Routes to Conversion or Reuse of Recycled Ammonium Perchlorate

New, potentially green, and efficient synthetic routes for the remediation and/or re-use of perchlorate-based energetic materials have been developed. Four simple organic imidazolium- and phosphonium-based perchlorate salts/ionic liquids have been synthesized by simple, inexpensive, and nonhazardous methods, using ammonium perchlorate as the perchlorate source. By appropriate choice of the cation, perchlorate can be incorporated into an ionic liquid which serves as its own electrolyte for the electrochemical reduction of the perchlorate anion, allowing for the regeneration of the chloride-based parent ionic liquid. The electrochemical degradation of the hazardous perchlorate ion and its conversion to harmless chloride during electrolysis was studied using IR and Cl-35 NMR spectroscopies.