Faculty of Science School of Animal,Plant,Enviromental Sciences 0311595p NicrosM@social.mpu.gov.za / The antifolate combination of sulfadoxine and pyrimethamine (SP) is one of few remaining affordable drug combinations available for wide-scale treatment of uncomplicated Plasmodium falciparum malaria in Africa. In vivo studies of SP efficacy conducted during 1998, 2000 and 2002 at the Naas sentinel site in Mpumalanga province, South Africa, demonstrated a gradual non-significant increase in late treatment failure (LTF) and early treatment failure (ETF) resistance to SP, while gametocyte carriage increased significantly between 1998 and 2002 (p < 0.0001). This study aimed to determined and compare the frequency of dihydrofolate reductase (dhfr) and dihydropteroate synthetase (dhps) resistant haplotypes in P. falciparum parasites from patients treated with SP in three consecutive standardized in vivo therapeutic efficacy studies in Mpumalanga province, since implementation of SP as first line treatment in 1998, and to investigate associations between the presence of mutations and treatment outcomes after SP treatment. Four hundred-and-three samples were studied and 358 yielded polymerase chain reaction products. A novel high throughput sequence-specific oligonucleotide probe-based approach was used to examine the resistance status of the three in vivo P. falciparum populations. Screening for the presence of all known point mutations in dhfr and dhps genes revealed that only five dhfr and three dhps allelic haplotypes were present. In all the samples investigated, point mutations were identified only at codons 108, 51 and 59 of the dhfr gene and at codons 347 and 540 of the dhps gene. The prevalence of dhfr resistant haplotypes was 35.4% in 1998, 38.7% in 2000, and 41.0% in 2002, while the prevalence of dhps resistant haplotypes was 9.7% in 1998, 7.2% in 2000 and 41.6% in 2002, the latter representing a significant increase (p < 0.002). The prevalence in both dhfr and dhps gene resistant haplotypes were selected gradually during the three in vivo studies in Mpumalanga province. Infection with parasites having triple dhfr mutations and double dhps mutations, "the quintuple mutant", was associated with SP treatment failure (p < 0.001). Mutations at both dhfr and dhps loci may be important predictors of SP resistance in Mpumalanga province.

This research is inserted in the topic of timber buildings. Many construction systems are available for building using timber, with the two main systems in residential ambit being Cross Laminated Timber and Light Timber Frame. Both systems reckon on the presence of shear walls to bear the effects of horizontal loads like seismic events or wind. This thesis deals with timber shear walls, and is divided into two parts: the first part is related to the ultimate and serviceability limit states rules to be included in upcoming versions of the building codes, while the second part presents a novel use of CLT walls as seismic renovation for existing buildings, as part of a European project. The first part of the thesis, which is presented in three papers, is closely related to the process of producing new building codes, and aims at an easier integration between research and codification. The initial focus is the behaviour of Cross Laminated Timber subjected to in-plane loading. Eurocode 5 currently lacks a part concerning this product and the discussion is still ongoing regarding the methods for stresses evaluation and on the strength values to adopt for safety verifications. The first paper tackles this problem by analysing different calculation methods currently available for the evaluation of the in-plane shear stresses, a common notation is introduced in order to have a meaningful comparison between methods proposed by different authors. All methods are then applied to a real case of existing experimental data regarding a four point bending test of CLT beams. Stiffness and strength of CLT are essential parameters for the definition of models to be adopted in codes regarding timber buildings, in particular for the calculation of shear walls. Another very common timber construction system is called Light Timber Frame: an assembly comprising a timber frame and an external sheathing layer mechanically joined to the frame. Consequently LTF walls are considered, the study is directed towards shear wall models for the evaluation of deformations. The second paper focuses on the evaluation of the displacement at the top of LTF walls subjected to horizontal loads. This is a key aspect for designers, since the limitation of deformations ensures that the building retains a satisfactory performance at serviceability limit states. The displacement is due to many different contributions, with the sheathing-to-framing deformation being one of the major ones. The paper presents a comparison between two of the proposed methods to calculate the sheathing-to-framing deformation of LTF shear walls. The influence of the nail slip contribution on the overall displacement of the top of the wall is studied also with parametric analyses, by varying both mechanical properties and geometrical dimensions. Comparison with existing experimental data is also provided. The study on shear walls regards also their lateral capacity, as well as the comparison between LTF and CLT walls of equal aspect ratio and similar restraining. In the third paper, existing cyclic test data on LTF and CLT walls were used to study the different displacement contributions and estimate the influence of the hold-down on the lateral response of the walls. A simplified capacity model is proposed for the walls, based solely on the hold-down forces. The second part of the thesis deals with the use of CLT shear walls as a mean for the retrofit of existing buildings. The need for sustainable renovation solutions and improvement of the performance of existing buildings is at the base of the European project e-Safe. The project presents a multidisciplinary approach on building renovation, from mechanical, energetic, technological and architectural point of view. In this thesis the focus is on the seismic retrofit system called e-CLT: a CLT panel is attached to the outside of existing buildings with a novel connector that acts as a friction dissipation device, thus offering additional energy dissipation in case of strong earthquakes. The fourth paper presents the first experimental campaign on this novel friction connector. Different geometries for the connector are studied and optimised, before being tested under cyclic protocol. The connector is tested on a steel setup, in order to isolate the friction behaviour and study the stability of the hysteresis loops. The results permitted to acquire new information useful for further developments on the system. The fifth paper presents a subsequent experimental campaign on the friction connector. The shape is changed and improved in light of the previous results. The setup is improved and includes also a screw connection between friction connector and CLT panel. The goal is to study the influence of the timber connection on the friction dissipative performance. An analytical model is proposed, fitted on the experimental data.

The analysis of long-time formant distribution is relatively young but promising discipline of speaker identification. It is a method of mapping the long-term behavior of formants in speech of individual speakers. Frequently encountered problems in practice are bad acoustic quality and very short duration of analyzed recordings. This work aims to present the historical development of forensic phonetics and currently used methods. In the practical part, it deals with the usability of LTF method in forensic practice, especially in recordings containing background noise. It was shown that the noise appreciably affects extracted LTF values and unfortunately the change is not systematic. Therefore, we proposed several methods to compensate the noise in recordings, in order to be able to compare recordings with and without noise. We also investigated the minimum duration of recording, which is necessary for statistical reliability of the resulting values. This boundary is not exact and for each speaker, it is substantially individual. But it is apparent that recordings (vocalic streams) shorter than 15 s often provide incomplete information, wherefore they cannot be recommended for analysis. Keywords: LTF, long-time formant distribution, speaker identification, forensic phonetics, acoustic quality of...