Date & Place: September 28, 2021, 2:00pm-3:00pm, SA311.
Speaker: Prof. Ting-Wen Chen (陳亭妏). Institute of Bioinformatics and Systems Biology, National Yang Ming Chiao Tung University.
Title: Identification of cancer prognostic biosignatures from multiomic data
Abstract: 透過多體學整合分析台灣口腔癌病患的基因體及轉錄體資料，我們找出癌細胞基因體中的突變以及轉錄體內表現量改變的反應途徑。與開放資料庫上的資料比較更台灣與美國口腔癌病人在突變與反應途徑改變都非常的相似，不過台灣口腔癌病人中，特別常見APOBEC3B基因缺失多型性。更進一步分析發現這群帶有APOBEC3B基因缺失多型性的口腔癌病人當中若同時結合有APOBEC3A低表現量，則對現有的治療方式反應不良，這些發現提供了第一個基因多型性的預後生物標誌，並且顯示了APOBEC基因具有做為將口腔癌病人分群並實現精準醫療生物標誌的潛力。隨後我們應用相同的分析策略分析了開放癌症資料庫內的資料並找尋具有預後預測能力之生物標誌。

Date & Place: October 5, 2021, 2:00pm-3:00pm, SA311.
Speaker: Prof. Jin-Cheng Jiang (江金城). Department of Mathematics, National Tsing Hua University.
Title: On the small data Cauchy problem for the Boltzmann equation
Abstract: I will review the results concerning the small data Cauchy problem for the Boltzmann equation and show the progress on going.

Date & Place: October 12, 2021, 2:00pm-3:00pm, SA223.
Speaker: Prof. Shih-Hsien Yu (尤釋賢), Institutue of Mathematics, Academia Sinica.
Title: A computable approach to wave and heat equation
Abstract: In this talk we will give an elementary method to study wave and heat equation including various physical condition imposed.

Date & Place: October 19, 2021, 13:20pm-3:10pm, SA223.
Speaker: 孔守謙諮商心理師（拾光心理諮商所）
Title: 精神疾病認識與壓力調適

Date & Place: October 26, 2021, 2:00pm-3:00pm, SA223.
Speaker: Dr. Chun-Wei Chang (張俊偉). National Center for Theoretical Sciences.
Title: Reconstructing large interaction networks from empirical time series data
Abstract: Reconstructing interactions from observational data is a critical need for investigating natural biological networks, wherein network dimensionality is usually high. However, these pose a challenge to existing methods that can quantify only small interaction networks. Here, we proposed a novel approach to reconstruct high-dimensional interaction Jacobian networks using empirical time series without specific model assumptions. This method, named “multiview distance regularized S-map”, generalized the state space reconstruction to accommoDate high dimensionality and overcome difficulties in quantifying massive interactions with limited data. When evaluating this method using time series generated from theoretical models involving hundreds of interacting species, estimated strengths of interaction Jacobians were in good agreement with theoretical expectations. Applying this method to a natural bacterial community helped identify important species from the interaction network and revealed mechanisms governing the dynamical stability of a bacterial community. The proposed method overcame the challenge of high dimensionality in large natural dynamical systems.

Mideterm week (November 1-5, 2021) break.

Date: November 9, 2021, 2:00pm-3:00pm, SA223.
Speaker: Dr. Yu-Chan Chang (張育展). Emory University, Oxford College.
Title: Some interplay between graphs and groups
Abstract: Many interesting groups are defined on graphs, and their algebraic structures can be detected from the combinatorics of graphs. For example, right-angled Artin groups (RAAGs) are defined on finite simple graphs, and two RAAGs are isomorphic if and only if their defining graphs are isomorphic. RAAGs appear as subgroups in some important groups, such as mapping class groups and right-angled Coxeter groups. On the other hand, RAAGs contain many interesting subgroups, including 3-manifold groups and graph braid groups. Bestvina–Brady groups (BBGs) are exotic subgroups of RAAGs and have connections with group theory and topology. In the former, BBGs provide examples of groups that satisfy certain finiteness properties but not others. In the latter, there are BBGs that are either counterexamples to the Eilenberg–Ganea Conjecture or the Whitehead Conjecture. In this talk, I will survey some interplay between the algebraic structures of RAAGs/BBGs and the combinatorics of the defining graphs. Along the way, I will discuss the Dehn functions of these groups. If time permits, I will talk about a similarity between BBGs and matroids. Most of the talk will be accessible to graduate students.

Date: November 16, 2021, SA223.
Speaker: Prof. Chien-Ling Lo (駱建陵). College of Management, Yuan Ze University.
Title: Valuation of Financial Derivatives
Abstract: This study proposes an efficient approach for the pricing of VIX derivatives under the affine framework and investigates the respective value of two variance components and variance jumps in the pricing of VIX derivatives. Our numerical results show that our approach significantly reduce the computational burden. Our empirical findings provide support for the use of two-variance component models as the means of capturing the fickle term structure of VIX derivatives, and the use of variance jumps is vital when included in the long-run variance component.

Date: November 23, 2021.
Speaker: Prof. Kung-Chien Wu (吳恭儉). Department of Mathematics, National Cheng Kung University.
Title: Mixture estimate in fractional sense
Abstract: Mixture estimate arise in the study of well-posedness of solutions to the Boltzmann equations. This property says that some space regularity result will be obtained if one mixes the transport operator and an integral operator with velocity smoothing effect. In this talk we will present this property in fractional sense and apply it to the Boltzmann equation for very soft potentials.

Date: November 30, 2021.
Speaker: Dr. Liren Lin (林立人). Department of Applied Mathematics, National Sun Yat-sen University.
Title: Tangential locally linear embedding
Abstract: Hessian locally linear embedding (HLLE) is a manifold learning method which exhibits prominent performance for the “Swiss roll with a hole”. Its original idea however is sophisticated and abstruse. We will introduce a new interpretation of its mechanism. Based on this new understanding, we then propose a new method called tangential locally linear embedding. It is much simpler and more robust than HLLE.

Date: December 7, 2021
Speaker: Prof. Guan-Ru Yu (余冠儒). Department of Mathematics, National Kaohsiung Normal University.
Title: On the study of phylogenetic networks
Abstract: Phylogenetic networks have replaced phylogenetic trees in many applications in biology whitin past few decades. Biologists are particularly interested in some classes of phylogenetic networks. In this talk, I will present our recent work of counting the major classes of phylogenetic networks which are believed to be of great help to the study of phylogenetics. Moreover, I will also outline some open problems which we plan to handle in the near future.

Date: December 14, 2021
Speaker: Prof. Weichung Wang (王偉仲). Institute of Applied Mathematical Sciences, National Taiwan University.
Title: Building AI-Assisted Clinical Workflows via Medical Data Analytics
Abstract: The rapid development of Artificial Intelligence (AI) introduces a promising new era of smart and precision medicine. Interdisciplinary collaborations play a critical role in medical AI research, education, and productization. We will illustrate how the Medical Data Analytics Framework can achieve the end-to-end R&D life-cycle. The framework includes project design, multimodality data, intelligent analytics, medical workflows, regulation and ethics, and solution landing. We will also demonstrate how medical AI algorithms and tools can reduce physicians' loading and improve patient outcomes.

Date: December 21, 2021.
Speaker: Prof. Yi-Hsueh Brad Chuang (莊易學). Institute of Environmental Engineering, National Yang Ming Chiao Tung University.
Title: Development of novel UV-based advanced oxidation processes for water treatments
Abstract: In this seminar talk, Dr. Chuang will be sharing his research on the developments of novel photochemical advanced oxidation processes that he has done previously and has been doing. Dr. Chuang’s research focused on wastewater potable reuse, driven by the severe water scarcity in California, when he was a postdoc researcher. The severe water scarcity issues facilitate the use of highly treated domestic wastewater as an alternative source for drinking water in many states of the U.S. Full Advanced Treatment (FAT) trains consisting of microfiltration, reverse osmosis, and ultraviolet/hydrogen peroxide advanced oxidation process (UV/H2O2 AOP) are commonly employed for treating wastewater for potable reuse. Among these treatment units, the UV/H2O2 AOP represents a broad-screen chemical barrier by producing hydroxyl radical to destroy contaminants passing through RO. Moreover, utilities commonly add chloramines to control membrane fouling, and chloramines tend to pass through RO unit and appear in the RO permeate, making the subsequent UV/H2O2 AOP de facto UV/H2O2 + chloramines AOP. Dr. Chuang’s work revealed that the presence of chloramines impaired the efficiency of AOP in terms of micropollutant degradation rates. To address this issue, Dr. Chuang’s previous work also sought to find alternative oxidants or alternative AOP schemes that could outperform the current status quo. Inspired by his previous research, Dr. Chuang has kicked off research projects of using LED-UVA coupled with different oxidants to improve the efficacy of AOPs on degrading micropollutants after he started his assistant professorship in NYCU. Dr. Chuang’s group successfully developed a mathematical kinetics model encompassing nearly 100 elementary reactions. In his talk, Dr. Chuang is going to show the audiences how a mathematical kinetics model can help engineers to optimize their designs

Final exam week (December 24-30, 2021) break.