CS PhD PS求修改~！1150字是不是太长了点呀

My ever-growing interest in pursuing a doctoral degree in computer science stems from a series of questions I asked myself in different stages of my life. The very first one was why a computer could possibly own great power to facilitate our daily life while it seemed to be nothing but a combination of silicon, metals and plastics. This question inspired me to choose computer science as my major in college where I worked hard on courses to seek the answer and participated in commercial and research projects to make real life contributions. These projects included a still in use hospital management system and a nationally funded robot control system etc. When I was about to choose my career path in my senior, the following questions: how life functions with such tremendous complexity; how it evolved to possess this capacity and more importantly: how the computer as a tool can expedite the seeking of answers to the two previous questions ushered me the way becoming a computational biologist.

I then entered the MS-PhD program of bioinformatics of XXX University, Japan. Have prepared myself with classes and extracurricular readings, I soon got the chance to  conduct research in this field. In the first academic year, I visited Rice University as an exchange scholar and participated in a research project led by Dr. Luay Nakhleh. I defined the boundary of my work and carefully chose databases and programming languages to fit my purpose from a large reservoir before starting my work. In this project, I calculated correlation between gene duplicability (size of gene family) and gene complexity (number of domains and length of protein product, etc) as well as correlation between gene duplicability and connectivity (number of edges in PPI network) across Tree of Life. I for the first time discovered an increasing trend of the duplicability connectivity correlation which agreed with the increase of the genome size of respective species and also discovered that the duplicability complexity correlation were constant across species. My colleague ran computer simulations and proposed a neutral evolutionary explanation. Our study was published in PLOS ONE (Impact Factor 4.0) in 2012 with me as the second author. The successful first attempt encouraged me in my future adventure in this field.

With all the basic skills and understanding of research obtained from the past research experiences, I began to launch my own study to train myself to become an innovative and independent researcher. In the second academic year, I performed extensive literature review in molecular evolution especially in the context of gene duplication which I had examined in the previous study. I learned that the effect of ray-finned fish specific whole genome duplication (WGD) on coding sequences has been intensively studied, while how WGD affected noncoding flanking regions (NCFRs) evolutionarily remains unclear. After designing the study with my professor’s help based on this idea, I solved all problems with trials and errors before a meaningful result could be obtained. Because fish species with publicly available genomes are evolutionarily distant, I designed an algorithm to remove non-homologous sequences trying every possible parameter to get the most reliable alignments. The impact on mutation rate of CpG hypermutability and GC content were also addressed to avoid biased results. I also proposed a model to measure asymmetry of evolutionary rates.  Finally all sequence comparison and sorting related codes were iteratively optimized to handle the big sequence data more efficiently. In this study, I discovered the asymmetry of evolutionary rates of NCFRs of duplicated genes and that the asymmetry increased as the noncoding sites were closer to the coding sequences. I also confirmed selective constraints on NCFRs of orthologous genes in fish genomes, and found that they were much more remarkable than those in human genomes. A noncoding region version duplication-divergence explanation in which evolutionary forces influenced the gain and/or loss of regulatory elements in NCFRs and the large population size of fish were used to account for these results. These results served to deepen our understanding of evolution of duplicated genes and led to a to-be-submitted manuscript to Genome Biology and Evolution (Impact Factor 4.8).

However, my life in Japan is not all that smooth. My Japanese was not adequate for me to fully utilize the academic resources. The courses and exams were all given in Japanese, I only strived to get a fair performance as research have occupied most of my time. More importantly, based on my experience in Japan and US, I find the cultural characters of United States (e.g. openness rather than conservativeness, evaluation based on competency rather than hierarchy) better motivates young people to fully engage themselves to new challenges.  So, I quitted my doctoral program in Japan and made up my mind to pursue the PhD degree in the United States.

Aware that I have been concentrating on the biology and evolution questions, I am considering to address the computational question more closely as I gradually realized that the highly efficient and reliable bioinformatics tools are critical to downstream studies. University of Wisconsin’s computer science program is ideally suitable to me because it includes outstanding faculties devoted to devise such sophisticated tools. Dr. AAA developed state-of-art methods to quantify gene and isoform expression (RSEM) and to infer splice graphs using RNA-seq data. He also proposed a novel method to detect transcription factor binding sites (TFBSs) using multi-reads from ChIP-seq data. His work was specifically interesting to me because the correct inference of alternative splicing is critical to deriving reliable NCFR for each transcript in my research and TFBSs along with other regulatory elements were responsible for the evolutionary patterns I discovered. Dr. BBB`s research work also attracts me a lot. He developed a new data processing system WHAM to greatly accelerate genome alignment which was the most time-consuming part of all calculations in my research. I am also very interested in research works by Dr. CCC and Dr. DDD, because they try to address yet another fundamental question: what computer scientists could do to handle medical issues. They developed machine learning methods and various statistical and computational models and applied them to biomedical studies.  For example, Dr. CCC proposed a Bayesian network model to predict breast cancer risks using genetic variants from GWA studies and mammographic findings. Dr. DDD developed a demand-driven clustering method to predict adverse reaction to a medication using EMR data. Working with them could provide me with the opportunities to accomplish this transition. With solid background in computer science and sufficient training in the field of bioinformatics, as well as clear research interests in the future, I do think I can succeed in your PhD program.  After obtaining my PhD degree, I will continue my lifelong seeking of the answers of all above mentioned questions as a researcher in academia.

谢谢大家~~~！

太长了..
影响因子不用提吧... 教授心里应该有数

我也是菜鸟，但是感觉倒数第二段不用说Japan不好吧，也没必要比较美国和日本，只用说自己很喜欢美国，喜欢美国文化，学术氛围等等就可以了吧? 个人感觉啊

q198800287 发表于 2013-10-21 16:08
太长了..
影响因子不用提吧... 教授心里应该有数

是呀，但我已经把什么gpa，本科水项目什么的都删了。不能再短了。。

q198800287 发表于 2013-10-21 16:08
太长了..
影响因子不用提吧... 教授心里应该有数

但是这些是偏生物信息的，可能cs别的方向的老师不知道这些杂志呢～

顺其自然123 发表于 2013-10-21 17:13
我也是菜鸟，但是感觉倒数第二段不用说Japan不好吧，也没必要比较美国和日本，只用说自己很喜欢美国，喜欢美 ...

嗯嗯，把昨天刚把对比删了，就写喜欢美国的文化什么的了～