Explore the vast range of titles available.

Though the title of my thesis infers a unifying theme via the application of machine learning, the two projects that form the bulk of my graduate degree are frankly more disparate than they are similar. Both endeavors provide novel methods to a field where ground truth is obscure and/or limited, and both apply machine learning techniques in their methodologies. Those similarities notwithstanding, the scientific domains, technical applications, experimental designs, and overall goals remain independent. While having a thesis comprised from two independent parts may not be conventional, this is, as they say, not a bug but a feature. Working within (and occasionally across) two research domains has helped me to acquire a diverse skillset and has provided me with a better and broader understanding of machine learning practices for scientific research.As this thesis is composed of two linked, but distinct projects, the abstract (and chapters) is divided in two. The first section details work related to large-scale predictions of purchasable chemical space, and the second summarizes a novel method for automating diagnosis of melanocytic atypia in human histopathological samples.I. Large-Scale Predictions for Purchasable Chemical SpaceThere are now over 400 million compounds one can easily purchase from the ZINC database (zinc.docking.org). About 350 million (85%) of these compounds are affordable enough for the average academic lab to conduct a ligand discovery project. However, the molecular targets (proteins) that these purchasable compounds bind and modulate—if any—are rarely known. Fewer than 1 million compounds (<0.25%) have been reported active in a target-specific assay according to public databases such as ChEMBL. In the absence of target activity information, the process of selecting compounds for general purpose screening will often be target-naïve.To facilitate access to new chemistry for biology, my collaborator John Irwin and I generated predictions for all purchasable compounds in ZINC at the time. I explored methods for optimizing predictive performance of compound-target associations using ChEMBL’s bioactivity dataset (version 21) as a benchmark. Comparisons on cross-validation sets of the bioactivity dataset against several methods such as multinomial naïve-bayesion classifiers revealed that the combination of the Similarity Ensemble Approach (SEA) with the maximum Tanimoto similarity to the nearest bioactive yielded the best performance. I verified the utility for several of these predictions, quantified target prediction biases inherent to the dataset, and provided thresholding suggestions to the user for controlling sensitivity and specificity of the predictions, as well as novelty of target-associations allowed.II. Automating Diagnosis of Melanocytic Atypia: A Precursor to Melanoma in SituMelanocytic atypia, a biological precursor to Melanoma, is histopathologically challenging. Pathologist interobserver agreement for melanocytic atypia in standard (H&E) histology images is low, ranging from 33-68%, with melanoma in situ (MIS) in particular contributing to diagnostic discordance. A lack of agreement among experts presents a challenge to any supervised learning task, where the utility of a learned function depends on the accuracy and reliability of labels used.To circumvent the issue of pathological discordance in labeling, I paired H&E histology images with contiguously cut tissue sections, immunohistochemically (IHC) stained for melanocytes. I developed a deep-learning pipeline for automating diagnosis of melanocytic atypia using this custom dataset of paired, whole slide images (WSIs) and trained convolutional neural networks to identify the presence of melanocytes in H&E sections, using information solely from paired images. Networks achieved strong performance on holdout patient datasets. For each network trained, I generated full-scale (20X magnification) high-resolution (pixel-wise) prediction heatmaps on holdout tissue sections (H&E), for pathological interpretation, and applied saliency mapping to show what networks attend to in H&E images. This pipeline aims to provide assistance to the clinical pathologist to reach better consensus regarding new MIS diagnoses in cutaneous biopsies.

Ubiquitin is essential for eukaryotic life and varies in only 3 amino acid positions between yeast and humans. However, recent deep sequencing studies indicate that ubiquitin is highly tolerant to single mutations. We hypothesized that this tolerance would be reduced by chemically induced physiologic perturbations. To test this hypothesis, a class of first year UCSF graduate students employed deep mutational scanning to determine the fitness landscape of all possible single residue mutations in the presence of five different small molecule perturbations. These perturbations uncover 'shared sensitized positions' localized to areas around the hydrophobic patch and the C-terminus. In addition, we identified perturbation specific effects such as a sensitization of His68 in HU and a tolerance to mutation at Lys63 in DTT. Our data show how chemical stresses can reduce buffering effects in the ubiquitin proteasome system. Finally, this study demonstrates the potential of lab-based interdisciplinary graduate curriculum. The ability of an organism to grow and reproduce, that is, it’s “fitness”, is determined by how its genes interact with the environment. Yeast is a model organism in which researchers can control the exact mutations present in the yeast’s genes (its genotype) and the conditions in which the yeast cells live (their environment). This allows researchers to measure how a yeast cell’s genotype and environment affect its fitness. Ubiquitin is a protein that many organisms depend on to manage cell stress by acting as a tag that targets other proteins for degradation. Essential proteins such as ubiquitin often remain unchanged by mutation over long periods of time. As a result, these proteins evolve very slowly. Like all proteins, ubiquitin is built from a chain of amino acid molecules linked together, and the ubiquitin proteins of yeast and humans are made of almost identical sequences of amino acids. Although ubiquitin has barely changed its sequence over evolution, previous studies have shown that – under normal growth conditions in the laboratory – most amino acids in ubiquitin can be mutated without any loss of cell fitness. This led Mavor et al. to hypothesize that treating the yeast cells with chemicals that cause cell stress might lead to amino acids in ubiquitin becoming more sensitive to mutation. To test this idea, a class of graduate students at the University of California, San Francisco grew yeast cells with different ubiquitin mutations together, and with different chemicals that induce cell stress, and measured their growth rates. Sequencing the ubiquitin gene in the thousands of tested yeast cells revealed that three of the chemicals cause a shared set of amino acids in ubiquitin to become more sensitive to mutation. This result suggests that these amino acids are important for the stress response, possibly by altering the ability of yeast cells to target certain proteins for degradation. Conversely, another chemical causes yeast to become more tolerant to changes in the ubiquitin sequence. The experiments also link changes in particular amino acids in ubiquitin to specific stress responses. Mavor et al. show that many of ubquitin’s amino acids are sensitive to mutation under different stress conditions, while others can be mutated to form different amino acids without effecting fitness. By testing the effects of other chemicals, future experiments could further characterize how the yeast’s genotype and environment interact.

Optimal brain function depends upon efficient control over the brain entry of blood components; this is provided by the blood-brain barrier (BBB). Curiously, some brain-impermeable drugs can still cause behavioral side effects. To investigate this phenomenon, we asked whether the promiscuous drug efflux transporter Mdr1 has dual functions in transporting drugs and endogenous molecules. If this is true, brain-impermeable drugs may cause behavioral side effects by affecting brain levels of endogenous molecules. Using computational, genetic and pharmacologic approaches across diverse organisms we demonstrate that BBB-localized efflux transporters are critical for regulating brain levels of endogenous steroids, and steroid-regulated behaviors (sleep in Drosophila and anxiety in mice). Furthermore, we show that Mdr1-interacting drugs are associated with anxiety-related behaviors in humans. We propose a general mechanism for common behavioral side effects of prescription drugs: pharmacologically challenging BBB efflux transporters disrupts brain levels of endogenous substrates, and implicates the BBB in behavioral regulation.

Ubiquitination is an essential post-translational regulatory process that can control protein stability, localization, and activity. Ubiquitin is essential for eukaryotic life and is highly conserved, varying in only 3 amino acid positions between yeast and humans. However, recent deep sequencing studies in S. cerevisiae indicate that ubiquitin is highly tolerant to single amino acid mutations. To resolve this paradox, we hypothesized that the set of tolerated substitutions would be reduced when the cultures are not grown in rich media conditions and that chemically induced physiologic perturbations might unmask constraints on the ubiquitin sequence. To test this hypothesis, a class of first year UCSF graduate students employed a deep mutational scanning procedure to determine the fitness landscape of a library of all possible single amino acid mutations of ubiquitin in the presence of one of five small molecule perturbations: MG132, Dithiothreitol (DTT), Hydroxyurea (HU), Caffeine, and DMSO. Our data reveal that the number of tolerated substitutions is greatly reduced by DTT, HU, or Caffeine, and that these perturbations uncover shared sensitized positions localized to areas around the hydrophobic patch and to the C-terminus. We also show perturbation specific effects including the sensitization of His68 in HU and tolerance to mutation at Lys63 in DTT. Taken together, our data suggest that chemical stress reduces buffering effects in the ubiquitin proteasome system, revealing previously hidden fitness defects. By expanding the set of chemical perturbations assayed, potentially by other classroom-based experiences, we will be able to further address the apparent dichotomy between the extreme sequence conservation and the experimentally observed mutational tolerance of ubiquitin. Finally, this study demonstrates the realized potential of a project lab-based interdisciplinary graduate curriculum.

Background To successfully reduce the negative impacts of stroke, high-quality health and care practices are needed across the entire stroke care pathway. These practices are not always shared across organisations. Quality improvement collaboratives (QICs) offer a unique opportunity for key stakeholders from different organisations to share, learn and ‘take home’ best practice examples, to support local improvement efforts. This systematic review assessed the effectiveness of QICs in improving stroke care and explored the facilitators and barriers to implementing this approach. Methods Five electronic databases (MEDLINE, CINAHL, EMBASE, PsycINFO, and Cochrane Library) were searched up to June 2020, and reference lists of included studies and relevant reviews were screened. Studies conducted in an adult stroke care setting, which involved multi-professional stroke teams participating in a QIC, were included. Data was extracted by one reviewer and checked by a second. For overall effectiveness, a vote-counting method was used. Data regarding facilitators and barriers was extracted and mapped to the Consolidated Framework for Implementation Research (CFIR). Results Twenty papers describing twelve QICs used in stroke care were included. QICs varied in their setting, part of the stroke care pathway, and their improvement focus. QIC participation was associated with improvements in clinical processes, but improvements in patient and other outcomes were limited. Key facilitators were inter- and intra-organisational networking, feedback mechanisms, leadership engagement, and access to best practice examples. Key barriers were structural changes during the QIC’s active period, lack of organisational support or prioritisation of QIC activities, and insufficient time and resources to participate in QIC activities. Patient and carer involvement, and health inequalities, were rarely considered. Conclusions QICs are associated with improving clinical processes in stroke care; however, their short-term nature means uncertainty remains as to whether they benefit patient outcomes. Evidence around using a QIC to achieve system-level change in stroke is equivocal. QIC implementation can be influenced by individual and organisational level factors, and future efforts to improve stroke care using a QIC should be informed by the facilitators and barriers identified. Future research is needed to explore the sustainability of improvements when QIC support is withdrawn. Trial registration Protocol registered on PROSPERO ( CRD42020193966 ).

A polygalacturonase (PG) isozyme was isolated from Penicillium solitum-decayed Anjou pear fruit and purified to homogeneity with a multistep process. Both gel filtration and cation exchange chromatography revealed a single PG activity peak, and analysis of the purified protein showed a single band with a molecular mass of 43 kDa, which is of fungal origin. The purified enzyme was active from pH 3.5-6, with an optimum at pH 4.5. PG activity was detectable 0-70 C with 50 C maximum. The purified isozyme was inhibited by the divalent cations Ca 2+ , Mg 2+ , Mn 2+ and Fe 2+ and analysis of enzymatic hydrolysis products revealed polygalacturonic acid monomers and oligomers. The purified enzyme has an isoelectric point of 5.3 and is not associated with a glycosylated protein. The PG isozyme macerated fruit tissue plugs in vitro and produced ~1.2-fold more soluble polyuronides from pear than from apple tissue, which further substantiates the role of PG in postharvest decay. Data from this study show for the first time that the purified PG produced in decayed Anjou pear by P. solitum, a weakly virulent fungus, is different from that PG produced by the same fungus in decayed apple.

A polygalacturonase (PG) isozyme was isolated from Penicillium solitum-decayed Anjou pear fruit and purified to homogeneity with a multistep process. Both gel filtration and cation exchange chromatography revealed a single PG activity peak, and analysis of the purified protein showed a single band with a molecular mass of 43 kDa, which is of fungal origin. The purified enzyme was active from pH 3.5–6, with an optimum at pH 4.5. PG activity was detectable 0–70 C with 50 C maximum. The purified isozyme was inhibited by the divalent cations Ca2+, Mg2+, Mn2+ and Fe2+ and analysis of enzymatic hydrolysis products revealed polygalacturonic acid monomers and oligomers. The purified enzyme has an isoelectric point of 5.3 and is not associated with a glycosylated protein. The PG isozyme macerated fruit tissue plugs in vitro and produced ,1.2-fold more soluble polyuronides from pear than from apple tissue, which further substantiates the role of PG in postharvest decay. Data from this study show for the first time that the purified PG produced in decayed Anjou pear by P. solitum, a weakly virulent fungus, is different from that PG produced by the same fungus in decayed apple.