完了！斯坦福博士（已拿到学位）已入职乡镇公务员
版主： Softfist
-
wanmeishijie（石昊）楼主
- 论坛元老
wanmeishijie 的博客 - 帖子互动： 2377
- 帖子： 72747
- 注册时间： 2022年 12月 10日 23:58
#31 Re: 完了！斯坦福博士（已拿到学位）已入职乡镇公务员
想想太雷，三年毕业
理解了老将是代入狗的视角之后，你就理解了老将
viewtopic.php?t=120513
理解了它们是代入狗的视角之后，它们为什么会嘲笑不愿意当狗的人，以及为什么会害怕想要反抗的人，就都可以理解了：
“放着好好的狗不当”
#39 Re: 完了！斯坦福博士（已拿到学位）已入职乡镇公务员
博士论文致谢部分
Acknowledgments
The journey of a PhD is a transition from a student to a scientific researcher, from knowledge receiving to contributing. Many people have told me that what really matters in a PhD is how much I can contribute to the scientific world. However, when I look back to the previous 6 years, what comes to my mind first is not what I have contributed, but all the kind help I have received from my advisors, mentors, group members, friends, and family. It is impossible for me to complete this journey alone without their support.
First, I want to thank my advisor Professor Mike Dunne for his guidance and support. No matter how busy he is, Mike always finds some time to teach me how to perform a critical study, how to apply my conceptual work into realistic applications, how to make a scientific report, how to deal with stresses in life ... Whenever I am confused about what to focus next, Mike can always guide me to the right direction with his expertise in various scientific fields. When I lose confidence in my research especially in the first a few years, Mike always encourages me with patience and appreciates all my work. Mike helps me in my study, research, and life. My gratitude to him is beyond words. I also want to thank my co-advisor Professor David Reis, thesis reader Professor Britt Hedman, and other dissertation committee members: Professor Gordon Wetzstein, Professor Soichi Wakatsuki, and Professor Zhirong Huang. They give me so many valuable suggestions to broaden and deepen my research topics from their expertise.
I also want to send special thanks to my mentor Dr. Chun Hong Yoon, who literally helps me every day in 6 years. We discuss everything from the history of serial X-ray crystallography, to the most frontier machine learning method in structural biology. He can always come up with interesting ideas that are worth trying, and he has so much knowledge in a variety of scientific fields from physics, biology, to computer science. Even when both of us get confused by some problems, he can always find the experts in those fields from his friend list. He helps tremendously not only in science, but also the life as a PhD candidate. During this pandemic, we still discuss the details of my research online despite his busy schedule. I really thank him for his great help through my PhD journey. He truly deserves half of all my work.
Then, I also want to thank Professor Mark Wilson and Dr. Michael E. Wall. They are experts in protein diffuse scattering, and have helped me a lot in the theory, experiment, and manuscript preparation of our diffuse scattering work. Dr. Daniel DePonte is the expert in serial crystallography experiment, who has helped me so much in our work of automated drop-on-demand sample delivery system. Without their help, it is impossible for me to get our study published so quickly.
I also want to thank all previous and current group members. They all helped me a lot during my PhD. Dr. Hsing-Yin Chang helped me in simulating X-ray diffraction data; Dr. Ariana Peck and Dr. Fr´ed´eric Poitevin helped me a lot in my diffuse scattering work. I also have many discussions in X-ray crystallography with Dr. Antoine Dujardin, Dr. Marc Griesemer, Axel Levy, Dr. Haoyuan Li, Dr. Monarin Uervirojnangkoorn, Dr. Cong Wang, Chen Wu, and David Zhang. I also want to thank the beamline scientist Dr. Raymond Sierra, who helped me a lot in many experiments and my jitter correction study.
I want to express my gratitude to the student office staffs in the department of Applied Physics, Claire Nicholas, Fiona Chiu, Patrice O’Dwyer, and Paula Perron. They always give me the most accurate and instant help through my PhD life.
I want to thank my family, my parents, grandparents, and my sister. My grandparents are about 90 years old, they always encourage me to follow my heart. My parents and my sister always support me, talk with me and help me to get rid of pressures in life. We really miss each other during this pandemic and I hope they can stay safe and healthy. Even separated far away from each other, I believe our feelings can be shared by the sun and moon every day and night.
Acknowledgments
The journey of a PhD is a transition from a student to a scientific researcher, from knowledge receiving to contributing. Many people have told me that what really matters in a PhD is how much I can contribute to the scientific world. However, when I look back to the previous 6 years, what comes to my mind first is not what I have contributed, but all the kind help I have received from my advisors, mentors, group members, friends, and family. It is impossible for me to complete this journey alone without their support.
First, I want to thank my advisor Professor Mike Dunne for his guidance and support. No matter how busy he is, Mike always finds some time to teach me how to perform a critical study, how to apply my conceptual work into realistic applications, how to make a scientific report, how to deal with stresses in life ... Whenever I am confused about what to focus next, Mike can always guide me to the right direction with his expertise in various scientific fields. When I lose confidence in my research especially in the first a few years, Mike always encourages me with patience and appreciates all my work. Mike helps me in my study, research, and life. My gratitude to him is beyond words. I also want to thank my co-advisor Professor David Reis, thesis reader Professor Britt Hedman, and other dissertation committee members: Professor Gordon Wetzstein, Professor Soichi Wakatsuki, and Professor Zhirong Huang. They give me so many valuable suggestions to broaden and deepen my research topics from their expertise.
I also want to send special thanks to my mentor Dr. Chun Hong Yoon, who literally helps me every day in 6 years. We discuss everything from the history of serial X-ray crystallography, to the most frontier machine learning method in structural biology. He can always come up with interesting ideas that are worth trying, and he has so much knowledge in a variety of scientific fields from physics, biology, to computer science. Even when both of us get confused by some problems, he can always find the experts in those fields from his friend list. He helps tremendously not only in science, but also the life as a PhD candidate. During this pandemic, we still discuss the details of my research online despite his busy schedule. I really thank him for his great help through my PhD journey. He truly deserves half of all my work.
Then, I also want to thank Professor Mark Wilson and Dr. Michael E. Wall. They are experts in protein diffuse scattering, and have helped me a lot in the theory, experiment, and manuscript preparation of our diffuse scattering work. Dr. Daniel DePonte is the expert in serial crystallography experiment, who has helped me so much in our work of automated drop-on-demand sample delivery system. Without their help, it is impossible for me to get our study published so quickly.
I also want to thank all previous and current group members. They all helped me a lot during my PhD. Dr. Hsing-Yin Chang helped me in simulating X-ray diffraction data; Dr. Ariana Peck and Dr. Fr´ed´eric Poitevin helped me a lot in my diffuse scattering work. I also have many discussions in X-ray crystallography with Dr. Antoine Dujardin, Dr. Marc Griesemer, Axel Levy, Dr. Haoyuan Li, Dr. Monarin Uervirojnangkoorn, Dr. Cong Wang, Chen Wu, and David Zhang. I also want to thank the beamline scientist Dr. Raymond Sierra, who helped me a lot in many experiments and my jitter correction study.
I want to express my gratitude to the student office staffs in the department of Applied Physics, Claire Nicholas, Fiona Chiu, Patrice O’Dwyer, and Paula Perron. They always give me the most accurate and instant help through my PhD life.
I want to thank my family, my parents, grandparents, and my sister. My grandparents are about 90 years old, they always encourage me to follow my heart. My parents and my sister always support me, talk with me and help me to get rid of pressures in life. We really miss each other during this pandemic and I hope they can stay safe and healthy. Even separated far away from each other, I believe our feelings can be shared by the sun and moon every day and night.
#40 Re: 完了！斯坦福博士（已拿到学位）已入职乡镇公务员
博士论文摘要：
Abstract
Over its long history, macromolecular X-ray crystallography has proven to be the most popular method for structural biology. Recent advent of serial X-ray crystallography at X-ray free-electron laser facilities and synchrotrons have revolutionized the field allowing scientists to study proteins that were previously inaccessible. Radiation damage-free structures from microcrystals at room temperature are routinely studied these days. The next frontier of serial X-ray crystallography will be in studying the function of the proteins by reconstructing a detailed movie of the protein in motion. Conformational variations/changes of proteins are studied using diffuse scattering and pump-probe type of experiments. My research is focused on developing advanced data analysis methods in the field of macromolecular X-ray crystallography to extract more information from massive datasets and push the boundaries of the status quo. Here is a summary of projects I would like to highlight.
The first project advances the analysis of pump-probe and mixing-jet experiments by applying machine learning to correct for uncertainty/inaccuracy of measured pumping or mixing times in serial femtosecond X-ray crystallography (SFX) experiments. In pump-probe or mixing-jet experiments, we either excite the crystals using an optical light or mix the crystals with a catalyst to trigger motion, followed by an X-ray probe after a certain delay. Although these methods have been broadly used to study protein dynamics, few people have considered the effects of jitter in these inaccurate timestamps, which can be critical in the study of ultrafast protein dynamics. In our work, we have applied the diffusion map method to embed high-dimensional diffraction images into low-dimensional latent space to correct for the time jitter.
The second project is a reproducibility study of diffuse scattering in crystalline isocyanide hydratase in three forms, its wild type and two mutants. Different from ultrabright Bragg peaks that arise from the coherent diffraction of periodic crystal lattices, diffuse scattering is much weaker and is induced by imperfections inside the crystal lattice, such as protein motion and/or other forms of disorders. I have developed advanced methods to extract much weaker diffuse scattering signals from diffraction images, and then studied possible protein motions that cause these diffuse signals, by comparing liquid-like motions model to the experimental data. This work demonstrated a po-tential utility for selecting a preferred atomic displacement parameter model from diffuse scattering data.
The third project involves a proof-of-principle experimental validation of an automated sample delivery robot for SFX experiments. The automated drop on demand system can accelerate sample exchange for SFX experiments, eliminate sample contamination, and has the potential to greatly improve the efficiency of both sample and beam time usage at pulsed X-ray facilities. This new sample delivery technique will enable structure determination of proteins that are difficult to crystallize in large quantities.
Abstract
Over its long history, macromolecular X-ray crystallography has proven to be the most popular method for structural biology. Recent advent of serial X-ray crystallography at X-ray free-electron laser facilities and synchrotrons have revolutionized the field allowing scientists to study proteins that were previously inaccessible. Radiation damage-free structures from microcrystals at room temperature are routinely studied these days. The next frontier of serial X-ray crystallography will be in studying the function of the proteins by reconstructing a detailed movie of the protein in motion. Conformational variations/changes of proteins are studied using diffuse scattering and pump-probe type of experiments. My research is focused on developing advanced data analysis methods in the field of macromolecular X-ray crystallography to extract more information from massive datasets and push the boundaries of the status quo. Here is a summary of projects I would like to highlight.
The first project advances the analysis of pump-probe and mixing-jet experiments by applying machine learning to correct for uncertainty/inaccuracy of measured pumping or mixing times in serial femtosecond X-ray crystallography (SFX) experiments. In pump-probe or mixing-jet experiments, we either excite the crystals using an optical light or mix the crystals with a catalyst to trigger motion, followed by an X-ray probe after a certain delay. Although these methods have been broadly used to study protein dynamics, few people have considered the effects of jitter in these inaccurate timestamps, which can be critical in the study of ultrafast protein dynamics. In our work, we have applied the diffusion map method to embed high-dimensional diffraction images into low-dimensional latent space to correct for the time jitter.
The second project is a reproducibility study of diffuse scattering in crystalline isocyanide hydratase in three forms, its wild type and two mutants. Different from ultrabright Bragg peaks that arise from the coherent diffraction of periodic crystal lattices, diffuse scattering is much weaker and is induced by imperfections inside the crystal lattice, such as protein motion and/or other forms of disorders. I have developed advanced methods to extract much weaker diffuse scattering signals from diffraction images, and then studied possible protein motions that cause these diffuse signals, by comparing liquid-like motions model to the experimental data. This work demonstrated a po-tential utility for selecting a preferred atomic displacement parameter model from diffuse scattering data.
The third project involves a proof-of-principle experimental validation of an automated sample delivery robot for SFX experiments. The automated drop on demand system can accelerate sample exchange for SFX experiments, eliminate sample contamination, and has the potential to greatly improve the efficiency of both sample and beam time usage at pulsed X-ray facilities. This new sample delivery technique will enable structure determination of proteins that are difficult to crystallize in large quantities.