Copied selected text to selection primary: The development of new pharmaceuticals is a long and ardous process that typically requires more than 10 years from target identification to approved drug. This process often relies on high throughput screening of molecular libraries. However, this is a costly and time-intensive approach and the selection of molecules to screen is not obvious, especially in relation to the size of chemical space, which has been estimated to consist of 10 60 compounds. To accelerate this exploration, molecules can be obtained from virtual chemical libraries and tested in-silico using molecular docking. Still, such methods are incapable of handling the increasingly colossal virtual libraries, currently reaching into the billions. As the libraries continue to expand, a pre-selection of compounds will be necessitated to allow accurate docking-predictions. This project aims to investigate whether the search for ligands in vast molecular libraries can be made more efficient with the aid of classifiers extended with the conformal prediction framework. This is also explored in conjunction with a fragment based approach, where information from smaller molecules are used to predict larger, lead-like molecules. The methods in this project are retrospectively tested with two clinically relevant G protein-coupled receptor targets, A 2A and D 2 . Both of these targets are involved in devastating disease, including Parkinson’s disease and cancer. The framework developed in this project has the capacity to reduce a chemical library of > 170 million tenfold, while retaining the 80 % of molecules scoring among the top 1 % of the entire library. Furthermore, it is also capable of finding known ligands. This will allow for reduction of ultra-large chemical libraries to manageable sizes, and will allow increased sampling of selected molecules. Moreover, the framework can be used as a modular extension on top of almost any classifier. The fragment-based approaches that were tested in this project performed unreliably and will be explored further.

Proteins are crucial for all living organisms and they are involved in many different processes. The function of a protein is tightly coupled to its structure, yet to determine the structure experimentally is both non-trivial and expensive. Computational methods that are able to predict the structure are often the only possibility to obtain structural information for a particular protein. Structure prediction has come a long way since its inception. More advanced algorithms, refined mathematics and statistical analysis and use of machine learning techniques have improved this field considerably. Making a large number of protein models is relatively fast. The process of identifying and separating correct from less correct models, from a large set of plausible models, is also known as model quality assessment. Critical Assessment of Techniques for Protein Structure Prediction (CASP) is an international experiment to assess the various methods for structure prediction of proteins. CASP has shown the improvements of these different methods in model quality assessment, structure prediction as well as better model building. In the two studies done in this thesis, I have improved the model quality assessment part of this structure prediction problem for globular proteins, as well as trained the first such method dedicated towards membrane proteins. The work has resulted in a much-improved version of our previous model quality assessment program ProQ, and in addition I have also developed the first model quality assessment program specifically tailored for membrane proteins. / <p>QC 20120313</p>

In recent years, the edible biomedicinal products called nutraceuticals have been becoming more popular among the pharmaceutical industries and the consumers. In the process of developing nutraceuticals, in silico approaches play an important role in structural elucidation, receptor-ligand interactions, drug designing etc., that critically help the laboratory experiments to avoid biological and financial risk. In this thesis, three nutraceuticals possessing antimicrobial and anticancer activities have been studied. Firstly, a tertiary structure was elucidated for a coagulant protein (MO2.1) of Moringa oleifera based on homology modeling and also studied its oligomerization that is believed to interfere with its medicinal properties. Secondly, the antimicrobial efficiency of a limonoid from neem tree called ‘azadirachtin’ was studied with a bacterial (Proteus mirabilis) detoxification agent, glutathione S-transferase, to propose it as a potent drug candidate for urinary tract infections. Thirdly, sequence specific binding activity was analyzed for a plant alkaloid called ‘palmatine’ for the purpose of developing intercalators in cancer therapy. Cumulatively, we have used in silico methods to propose the structure of an antimicrobial peptide and also to understand the interactions between protein and nucleic acids with these nutraceuticals. / <p>QC 20130531</p>

No description available.

No description available.

No description available.

Wastewater-based epidemiology (WBE) analyzes wastewater for the presence of biological and chemical substances to make public health conclusions. COVID-19 disease is caused by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) that infected individuals shed also in their feces, making WBE an alternative way to track SARS-CoV-2 in populations. There are many limitations to the detection and quantification of SARS-CoV-2 from wastewater, such as sample quality, storage conditions or viral concentration. This thesis aims to determine the extent of these limitations and the factors that contribute to them. Other viruses can help the measurements for example Bovine coronavirus (BCoV) can be spiked as a process surrogate, while Pepper mild mottle virus (PMMoV), a fecal biomarker is used to estimate the prevalence of SARS-CoV-2 infection. This study involved two distinct wastewater samples. For method comparison both samples were processed with two methods: virus concentration by electronegative (EN) filtration or direct RNA extraction method. From the RNA extracts RT-qPCR assays were performed to identify and quantify SARS-CoV-2, BCoV, and PMMoV. Based on the obtained cycle threshold (Ct) values, viral gene copy numbers and virus concentration of the original wastewater samples were calculated. Statistical tests were conducted to assess suggested hypothesizes and variations within the data. Results revealed differences in viral contents due to different sample qualities and as a result of freezing and thawing. Furthermore, different sample processing methods led to differences in quantification. In conclusion, improving analysis of SARS-CoV-2 in wastewater using methodologies with better detection efficiency leads to more reliable results.

Sepsis is a life-threatening condition that is caused by a dysregulated host response to infection. Timely detection of sepsis and antibiotic treatment is important for the patient’s recovery from sepsis. Usually, when sepsis is detected, immediate antibiotic treatment is started with broad-spectrum antibiotics as it takes time to determine the correct antibiotic susceptibility. To overcome this problem, next-generation sequencing is seen as one possible development in clinical diagnostics in the future. Automated bioinformatics pipelines could be used initially for surveillance purposes but eventually for rapid clinical diagnosis. Therefore, the results of 1928 Diagnostics, an automated pipeline for whole-genome sequencing (WGS) data analysis, were compared with the results of an in-house developed pipeline for manual data processing by analyzing sepsis-associated Escherichia coli (SEPEC) WGS data. The pipelines were compared by assessing their predicted antimicrobial resistance (AMR) genes, virulence genes and epidemiological relatedness. In addition, the predicted resistance genes were compared to phenotypic antimicrobial susceptibility testing (AST) data from the clinical microbiology laboratory. All the results obtained from the 1928 Diagnostics and in-house pipeline were similar but differed in the number of virulence/predicted AMR genes, AMR gene variants, detection of species and epidemiologically related E. coli samples. Moreover, the predicted AMR genes from both pipelines did not show a good overall relation to the phenotypic AST result. More studies are needed to make predictions of genes from the WGS analysis more reliable so that WGS analysis can be used as a diagnostics tool in clinical laboratories in the future.

DNA profiling with short tandem repeat data generated with massively parallel sequencing is associated with several challenges. FDSTools is an open-source software which applies correction models based on a reference database to correct DNA profiles. The correction models aim to provide an accurate representation of the true DNA profile and associated artefacts. Low analytical thresholds in FDSTools are suggested to improve detection of minor profiles in complex mixtures. The objective was to optimise FDSTools analysis for ForenSeq data, and to establish a Swedish reference database. The FDSTools analysis was subsequently compared to default analysis with the commercial Universal Analysis Software, and the likelihood ratio was evaluated. The FDSTools Library file was adapted for ForenSeq data. FASTQ files from single- and mixed-source samples were analysed with the software. The concordance between the software was assessed, and analytical thresholds in FDSTools were optimised. Likelihood ratios were calculated for sequencing- and capillary electrophoresis data to investigate the benefit of sequence level information. A reference database and correction models could not be generated, meaning that uncorrected data was used. The two software showed a 98.5% concordance. Disconcordance was caused by allele drop-out in heterozygous loci which implicated that certain markers may require individual interpretation. Lowering the analytical thresholds in FDSTools appeared to improve mixture deconvolution, but the lack of correction models obscured interpretation. Hence, without correction models optimial analytical thresholds could not be defined. Likelihood ratio based on sequencing data was not consistently higher compared to capillary electrophoresis, suggesting that sequence information is not always advantageous.

No description available.