Deep learning transcriptomic model for prediction of pan-drug chemotherapeutic sensitivity

The emergence of precision oncology approaches has begun to inform clinical decision-making in diagnostic, prognostic, and treatment contexts. High-throughput technology has enabled machine learning algorithms to use the molecular characteristics of tumors to generate personalized therapies. However, precision oncology studies have yet to develop a predictive biomarker incorporating pan-cancer gene expression profiles to stratify tumors into similar drug sensitivity profiles. Here we show that a neural network with ten hidden layers accurately classifies pancancer cell lines into two distinct chemotherapeutic response groups based on a pan-drug dataset with 89.0% accuracy (AUC = 0.904). Using unsupervised clustering algorithms, we found a cohort of cell line gene expression data from the Genomics of Drug Sensitivity in Cancer could be clustered into two response groups with significant differences in pan-drug chemotherapeutic sensitivity. After applying the Boruta feature selection algorithm to this dataset, a deep learning model was developed to predict chemotherapeutic response groups. The model’s high classification efficacy validates our hypothesis that cell lines with similar gene expression profiles present similar pan-drug chemotherapeutic sensitivity. This finding provides evidence for the potential use of similar combinatorial biomarkers to select potent candidate drugs that maximize therapeutic response and minimize the cytotoxic burden. Future investigations should aim to recursively subcluster cell lines within the response clusters defined in this study to provide a higher resolution of potential patient response to chemotherapeutics.

The necessity for carnivore experimentation in predicting the next zoonotic disease epidemic

Zoonoses are human infections or diseases caused by disease spillover from vertebrate animals to people [1]. Spillover is the movement of pathogens from their normal host to a novel species [2]; this can occur through bodily fluids, bites, food, water, or contact with surfaces where infected animals have travelled [3]. Although some zoonoses remain established within populations and primarily affect only one person per spillover (classified as enzootic zoonoses—e.g., rabies), others can be transmitted between people and result in localized, or even global outbreaks [4]. Zoonoses account for over 60% of infectious diseases in humans [4] and can be caused by viruses, parasites, bacteria, or fungi. Of these, viral zoonoses prove to be of greatest detriment to the public on a widespread scale, as they are responsible for numerous epidemics and pandemics, including severe acute respiratory syndrome (SARS), Middle East respiratory syndrome (MERS), and the novel coronavirus (COVID-19) [5-7]. Research has also been conducted on different taxonomic orders of species, such as Carnivora — placental animals which obtain nutrients from flesh — and their viral spillover risk [11].

Sex and age impact prevalence and symptoms of vasovagal syncope: Implications for accurate diagnosis

Syncope is characterized by the transient loss of consciousness followed by spontaneous recovery. The mechanism which underlies this condition is reduced blood flow to the brain [1]. Vasovagal syncope, often termed reflex syncope, is the most common type of syncope [1]. Vasovagal Syncope is caused by the abnormal autonomic reflex to certain stimuli such as pain, micturition/defecation, fear, seeing blood, etc., which results in vasodilation and often times, bradycardia [1].

Gaining the upper hand on COVID-19 misinformation

As the Coronavirus Disease 2019 (COVID-19) pandemic evolved, information about the virus also accumulated. However, accompanied by the quick emergence of factual information was an even greater abundance of false information. For example, by March 2020, videos containing non-factual information on COVID-19 accounted for over one-quarter of the most viewed videos on YouTube — greatly exceeding the popularity of factual videos released by governments and health professionals [1]. The World Health Organization declared this massive flux of misinformation surrounding COVID-19 an “infodemic”, where it is hard to distinguish between factual and non-factual information [2].

3D bioprinting in microgravity: An end to organ donor shortages?

Countless lives have been saved with the advent of modern organ transplantation. However, the current shortage of compatible organ donors is limiting the life-saving potential of transplantation. According to the United Network for Organ Sharing, approximately 20 patients die each day in the United States while waiting for a transplant [1]. The discrepancy between supply and demand of organ donors is accentuated by a fundamental ethical dilemma associated with deceased organ donation: one person must die so that another may live [2]. The current viewpoint considers the viability of 3D bioprinting in microgravity as a solution to organ donor shortages. Current alternatives to deceased organ donation, including xenotransplantation and other state-of-the-art bioprinting techniques, are reviewed and compared to bioprinting in microgravity. The limitations of bioprinting within Earth’s gravitational field are also discussed, revealing the need for further research.

The relationship between maritime phosphate pollution and socioeconomic wellbeing

Phosphate rock reserves are expected to deplete in the next 50-100 years, with the point of highest phosphorus production predicted to be in 2030. Phosphate, the base of many fertilizers, is a non-renewable resource. Ocean phosphate concentrations provide a good indication of global fertilizer use, since agricultural runoff often contributes to increases in ocean phosphate concentration. This study explores the relationship between the concentration of phosphate in a nation’s maritime borders and the nation’s score on the Social Progress Index. The study aims to link findings with possible approaches to help meet two of the United Nations’ (UN) Sustainable Development Goals: creating sustainable communities, and conserving and sustainably using oceans. Phosphate concentration data were acquired from the National Oceanic and Atmospheric Administration and compared with factors of social welfare from the 2018 Social Progress Index. For each nation on the Social Progress Index, the nation’s score on every factor was separately compared to ocean phosphate concentration data within that nation’s maritime borders, and a linear regression was performed for each comparison. The results indicate countries ranking higher on the Social Progress Index generally have greater ocean phosphate concentrations, suggesting that countries of higher social welfare contribute more to global phosphate use or have greater amounts of fertilizer runoff. The findings should be considered by developed nations to inform decisions around pollution reduction as well as developing nations aiming for sustainable social progress. Both should consider the environmental effects that social progress has or will have on the greater global community, of which the significance to sustainable resource development and environmental protection is invaluable.

Tocilizumab: A promising treatment for COVID-19 patients

A novel Coronavirus was identified in China as SARS-CoV-2 in December 2019 and was later declared a pandemic by the World Health Organization [1]. In most patients, COVID-19 presents with mild or moderate symptoms such as fever, cough, fatigue, and myalgia. However, around 10% of patients develop more severe symptoms such as pneumonia and multiorgan failure, which can result in death [1-2]. Until vaccines are widely available, the only way to reduce fatality rates is to treat affected patients with existing drugs and therapy options.

Hong Kong Student Science Project Competition 2021

The Hong Kong Student Science Project Competition (HKSSPC) promotes the interest in science and technology among youth, develops their creativity and critical thinking skills through an innov ative application of science and technology, and ignites their passions and career interests in these areas. This year, the HKSSPC was organized by The Hong Kong Federation of Youth Groups, the Education Bureau, the Hong Kong Science Museum, and the Hong Kong Science and Technology Parks Corporation. Furthermore, it was supported by the Innovation and Technology Commission and the Hong Kong Young Academic of Sciences. We extend our thanks to all these groups for making this year’s competition a success. STEM Fellowship collaborated with the HKSSPC Secretariat to provide youth from Hong Kong with the unique opportunity to submit their work in the STEM Fellowship Journal. This year’s theme was “Inspiration from Living - Innovation from Science” with an emphasis on United Nations’ Sustainable Development Goals. The broad scope of the competition allowed participants to submit their work in a variety of areas such as water pollution, nanoparticles, artificial intelligence systems, agriculture, environmental health, plastics, waste reduction, and many more. We are pleased to share the creativity and ambitious drive for research demonstrated by HKSSPC’s participants in these proceedings. We would like to congratulate every passionate individual who participated in the HKSSPC this year and wish them the best in their future STEM-related endeavours.

Using real-world transaction data to identify money laundering: Leveraging traditional regression and machine learning techniques

Money laundering is a pervasive legal and economic problem that hides criminal activity. Identifying money laundering is a priority for both banks and governments, thus, machine learning algorithms have emerged as a possible strategy to detect suspicious financial activity within financial institutions. We used traditional regression and supervised machine learning techniques to identify bank customers at an increased risk of committing money laundering. Specifically, we assessed whether model performance differed across varying operationalizations of the outcome (e.g., multinomial vs. binary classification) and determined whether the inclusion of investigator-derived novel features (e.g., averages across existing features) could improve model performance. We received two proprietary datasets from Scotiabank, a large bank headquartered in Canada. The datasets included customer account information (N = 4,469) and customers’ monthly transaction histories (N = 2,827) from April 15, 2019 to April 15, 2020. We implemented traditional logistic regression, logistic regression with LASSO regularization (LASSO), K-nearest neighbours (KNN), and extreme gradient boosted models (XGBoost). Results indicated that traditional logistic regression with a binary outcome, conducted with investigator-derived novel features, performed the best with an F1 score of 0.79 and accuracy of 0.72. Models with a binary outcome had higher accuracy than the multinomial models, but the F1 scores yielded mixed results. For KNN and XGBoost, we observed little change or worsening performance after the introduction of the investigator-derived novel features. However, the investigator-derived novel features improved model performance for LASSO and traditional logistic regression. Our findings demonstrate that investigators should consider different operationalizations of the outcome, where possible, and include novel features derived from existing features to potentially improve the detection of customer at risk of committing money laundering.

Undergraduate Engineering Research Day 2021

The University of Toronto’s Undergraduate Engineering Research Day (UnERD) is an annual conference aimed at providing an opportunity for undergraduate engineering students to showcase their research to industry professionals and fellow students, inspiring the exchange of innovative solutions across a wide breadth of global challenges. STEM Fellowship came together with the UnERD organizers to provide a unique opportunity to the engineering students to publish their work in our STEM Fellowship Journal. For all the variation between project themes, it remains that all submissions are of incredibly high quality. Every abstract is demonstrative of immense creativity and high potential on the respective team’s part. On behalf of STEM Fellowship, I would like to extend my heartfelt congratulations to all students who participated in UnERD, and I wish them all the best for their future endeavours in research and engineering. It has been a privilege for us to witness the research capabilities of the next generation of students firsthand, and I am certain all entrants will continue to demonstrate excellence in their respective research careers.