Browsing by Subject "Normalization"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
Item Developing next-generation methods for scoring CRISPR screening data(2022-04) Ward, HenryMapping the genotype-to-phenotype connection is a central goal of modern biology. CRISPR screening technology powers mapping efforts by identifying relationships between gene knockouts and conditions of interest, such as another gene knockout or the presence of a drug. Quantitative readouts from CRISPR screening experiments, however, are confounded by several different types of biases which impact the efficacy of genotype-to-phenotype mapping efforts. My contributions to the scientific community consist of several computational tools which address bias in CRISPR screening experiments. The first, Orthrus, addresses biases specific to combinatorial screening experiments which knock out multiple genes simultaneously. The second, Orobas, presents an improved scoring method for chemogenomic screening experiments and substantially reduces false positives compared to the existing state-of-the-art method. Lastly, I detail a novel technique called onion normalization for reducing technical bias in large-scale CRISPR screening experiments. Overall, these contributions serve as a methodological bedrock for the construction of improved genotype-to-phenotype maps from CRISPR screening data.Item The importance of being proportional: a paradigm shift for intensity-based label free relative quantification in mass spectrometry proteomics(2013-05) Van Riper, Susan KayeBiological variation not only provides insight into the molecular machinery of disease progression, but accurately informs clinicians about a patient's health status, both current and future. Researchers discover biological variation by conducting large scale comparative studies aimed at detecting differences in the molecular makeup (biomarkers) of samples in different states. Ideally suited for biomarkers are proteins because their cellular composition (proteome) and their degraded parts, endogenous peptides (peptidome), change in response to their environment and disease progression. For comparative proteomic studies, researchers commonly employ high performance liquid chromatography, coupled with electrospray ionization tandem mass spectrometry (HPLC-ESI-MS/MS) and labeled quantification. However, intensity-based label free relative quantification (iLFRQ) is more desirable than labeled quantification because iLFRQ is more cost effective and does not limit the number of samples in a study. Unfortunately, iLFRQ for proteins, and especially peptides, is challenging. Here, I highlight three challenges. 1) I contend that the current relative abundance paradigm is ill-suited to detect biological variation using iLFRQ. 2) HPLC-ESI-MS/ MS analyses produce poorly repeatable and reproducible results, and current normalization methods fail to mitigate localized extraneous variability (complex variability in measurements) from transient stochastic events occurring during an HPLC-ESI-MS/MS run. 3) Current software frameworks report protein level quantification rather than peptide level quantification. To overcome these challenges, I offer three contributions. 1) I propose to use the proportionality paradigm for iLFRQ instead of the relative abundance paradigm. 2) Proximity-based Intensity Normalization (PIN), an embodiment of the proportionality paradigm, normalizes a peptide's signal intensity by constructing its temporal neighborhood and computing its relative proportion within that neighborhood. 3) RIPPER, a new software framework that reports normalized peptide signal intensities rather than protein intensities. Evaluation results demonstrate that PIN dominates current normalization methods in reducing systematic bias and complex variability. Furthermore, RIPPER/PIN finds statistically significant biological variation which is now falsely reported or missed. I expect the proportionality paradigm for iLFRQ, embodied in PIN, and implemented in RIPPER, to change the way researchers analyze HPLC-ESI-MS/MS experimental data. The upshot will, I expect, will be reproducibility and repeatability improved, and otherwise falsely reported or missed, statistically significant biological variation discovered.Item A Simulation Study of Patient Accrual Patterns in Clinical Trials and Data Analysis of Histone 3 Lysine 36 Trimethylation ChIP-seq in Human Kidney Cancer(2017-07) Yu, ShichaoIn part one, we simulated a successive of two-armed randomized clinical trial with the time-to-event outcome over 15 years. We used three different accrual pattern representing slow, medium and fast accrual, which is in fact related to the number of trials for the sequential trials interested in the 15-year period. We used a historical survival distribution to explore the treatment effects and analyzed by the Cox proportional hazard ratio model and log-rank test. We computed the mean and median overall hazard ratio (year 15 versus year 0), and the probability of detrimental effect to find the optimal design parameters. Finally, we carried out a sensitivity analysis to study the effect of an additional 6 month turnaround time. In Part two, we have described a general workflow for the normalization of ChIP-seq data by estimating the normalization factor from peak-less regions. Using publicly available histone 3 lysine 36 trimethylation (H3K36me3) data from human kidney cancer, we demonstrated the better performance of our method over the existing approach.