MICROBIOLOGY AND MOLECULAR BIOLOGY

High School

Developing an Convolutional Neural Network for Accurate and Inexpensive Lung Cancer Diagnosis
Aishwarya Ayyanathan, American Heritage School Boca/Delray

Lung cancer is the leading cause of death by cancer around the world. As the American Cancer Society reports, almost 25% of all cancer deaths were caused by lung cancer. In 2020 alone, there is estimated to be a total of about 228,200 cases and 135,720 deaths by lung cancer in both men and women. One of the main causes of this high death rate is the difficulty of diagnosing it during early stages. Once the lung cancer reaches the late stages, when it is most commonly caught, it is often harder to treat and wrecks more havoc on the patient’s body. My project focused on applying Convolutional Neural Networks, a tool that analyzes visual imagery, to accurately diagnose malignant nodules in lung CT scan imagers. If an accuracy above the current threshold of 89% can be achieved, it would be a valuable aid for early diagnosis to both health care professionals and patients.

Using AntiSMASH to Compare Antimicrobial Genes of Commensal E.Coli (Normal Flora) to Pathogenic E.Coli
Chloe Barreto-Massad, American Heritage School Boca/Delray

Using AntiSMASH, a genetic software tool, I compared commensal vs. pathogenic E. Coli to determine which strain of bacteria was more likely to produce antibiotics in their lived habitat (soil).

New Computational Model For Mitosis To Save Lives From Cancer
Michael Batavia, The Bronx High School of Science

In this project, I create a method to individually pre-diagnose the many Americans affected with breast cancer using a custom convolutional neural network (CNN). Although the effects of breast cancer can be limited with treatments like chemotherapy, the cancer must be diagnosed before it is treated. Since early cancer diagnosis leads to a 100% survival rate over 5 years but later cancer diagnosis leads to only a 22% survival rate over 5 years, it is critical that breast cancer be diagnosed as early as possible before treatments become ineffective. By using a custom CNN, powered by multiple GPUs, hypertuning and parallelized data processing, my research presents a 83% accuracy in predicting the breast cancer pre-diagnosis of a patient from their lymph node tumors and a 89% score in dichotomizing between benign and cancerous tumors. These scores are 10% better than pathologist classification in both early and later stages of cancer. As a result of my research, other researchers can create similar CNNs to detect related cancers such as thyroid cancer and pancreatic cancer to pre-diagnose the increasing number of people diagnosed with these diseases worldwide.

Combatting Bacterial Biofilm: Generating a Non-toxic, Multi-pathway Targeted Cocktail Treatment Composed of N-Acetylcysteine, Carvacrol, and DNase to Inhibit Pseudomonas Biofilm Proliferation In Vitro
Shriya Bhat, Plano ISD

I developed an inexpensive and non-toxic multipathway targeted triple cocktail treatment composed of N-Acetylcysteine, Carvacrol, and DNase to target three main pathways of Pseudomonas biofilm formation, including pyocyanin, quorum sensing, and the eDNA-biofilm matrix.

Spaceflight Effects on Dermal Mechanical Strength: An Investigation into Collagen Production
Mira Chatrathi and Paolo Canigiula, Prospect High School and Lowell High School

Through the GeneLab for High Schools program at NASA Ames, we analyzed RNA-Sequencing data from mice that had been sent to the ISS, comparing the differentially expressed genes of mice in spaceflight and in ground control. We observed that in the spaceflight samples, there were several over-represented gene categories that pertained to collagen and the skin. We also found collagen genes, specifically COL5A3, to be up-regulated in spaceflight. Through analyzing this data and reading previous studies on spaceflight effects on the dermis, we hypothesize that the the up-regulation of Col5a3 in spaceflight and the alteration of genes involved in collagen assembly could contribute to decreased mechanical strength and increased permeability of the skin. Our research can help further studies find solutions to maintain dermal mechanical strength and reduce dermal permeability in space and on Earth, benefitting many people on and off the planet.

Analyzing Gut Microbiome and Microbial Metabolome to Identify Molecular Signatures for Early Detection of Pancreatic Ductal Adenocarcinoma
Leah Ellison, Pine Crest School

This research focuses on developing molecular signatures from analysis of gut microbiome and metabolome to eventually be used as diagnostic tools for Pancreatic Ductal Adenocarcinoma.

Characterizing the Role of Caffeine in Seizure Susceptibility and Duration in C. elegans
Mary Gabrielle, Pine Crest School

My research seeks to discover the role of different concentrations of caffeine on the duration of C. elegans recovery from seizures. This research utilizes low doses (10 and 30 mM) and high doses (60 and 90 mM) to target the mechanism of action that caffeine affects in C. elegans.

Machine Learning and Bioinformatics Approach Yields Noninvasive MiRNA Biomarkers for Early Lung Cancer Detection
Andrew Gao, Canyon Crest Academy

Non-small cell lung cancer kills millions of people but is very hard to diagnose at early stages when treatment is most effective. I used bioinformatics and machine learning analysis to identify reliable blood-based biomarkers for lung cancer diagnosis. I achieved 91.5% accuracy.

Identification and Analysis of Key Candidate Genes and Pathways in Lung Adenocarcinoma by Integrated Bioinformatics.
Prateek Gupta, American Heritage School Plantation

Using public databases and tools, I conducted an analysis to see what genetic factors lead to the development lung adenocarcinoma (LUAD). I created a list of differentially expressed, mutated, and environmentally responsive genes in LUAD and used it to find their protein-to-protein interactions, chemical interactions and whether alterations in those genes would reduce the survival rates of patients. Specific identification of genes can allow doctors and researchers to use them as targets for new treatment.

Genome Analysis of Photosynthetic Sea Slugs of the Elysia Genus
Grace Hancock, Pine Crest School

This presentation investigates the genetic causes of photosynthetic abilities in sacoglossan, or sea slugs, of the Elysia genus. The presentation is embedded within my presentation website.

Developing an Artificial Neural Network as a Method for Real-Time Diagnosis and Detection of Atrial Fibrillation in Electrocardiograms
Maryam Imran, American Heritage School Boca Delray

The purpose of this experiment was to develop an artificial neural network in order to accurately diagnose atrial fibrillation. Atrial fibrillation (AF or AFib) is a common type of arrhythmia that results in irregular electrical activity of the heart, in which the atria quiver due to disrupted signal transmission between the SA and AV nodes.

The Correlation Between Non-Coding RNAs and T Cell Exhaustion
Lulu Ingham, Pine Crest School

My research discusses the impact of T cell exhaustion and the promise microRNAs show as a way to combat it.

Correlation between Protein Type and Fatality Rates in Zaire ebolavirus
Ellen Jones, Haines City International Baccalaureate East

Establishing an association between proteins found in a strain of Zaire ebolavirus and the strain. I use an Analysis of Variance (ANOVA) to measure the correlation.

Using Deep Learning in Lyme Disease Diagnosis
Tejaswi Koduru, Thomas Jefferson High School for Science and Technology

This presentation will examine the process behind the creation of a machine learning algorithm to diagnose EM, or erythema migrans rashes, on patients with acute Lyme disease. The presentation will also examine the steps taken to implement this model into an app, allowing doctors and patients to rapidly diagnose the disease.

A Multi-OMICS Approach for Glaucoma
Ayesha Minhas, Pine Crest School

This research focuses on using the multi-OMICS approach for the identification of molecular therapies to treat glaucoma.

Improving Statistical Method BIO_CYCLE Performance by Training Process Optimization for Applications in Circadian Rhythm Pattern Detection
Roshni Mishra, American Heritage School Plantation

The circadian clock within organisms regulates essential functions over 24 hour cycles. Throughout the 24 hour cycle, different genes that regulate circadian rhythms called “clock genes” send signals to the suprachiasmatic nucleus (SCN) to control the various activities occurring in the body. In general, circadian rhythms are tied to the day and night cycle due to its relation to environmental cues such as lightness or darkness, exercise, social activity, and temperature. Due to this, a learning approach was developed in which machine learning was used to train deep neural networks and create BioCycle - the largest dataset containing biological data as well as synthetic data with time series. Remodeling the BIO_CYCLE code with preservations of the data correcting and manipulations that allow for correct representation of Circadian rhythm data, effectively increased the accuracy of NEW_CYCLE as demonstrated by the mean absolute error and mean absolute difference. It was hypothesized and supported that by making the code more up-to-date with simpler annotations, the neural network was clearer and better suited for both the biological and synthetic datasets. This demonstrates that by adapting the Gaussian process for 70/40 train/test ratio and creating 100 examples with 50,000 iterations in every mini batch, the accuracy of the neural network training method increased and is therefore a better statistical tool. Using a statistical tool to measure relatedness between synthetic and biological data allows for a cost-effective method as opposed to the laboratory methods of measuring omic data.

Testing the Extent of Synergistic Anti-Biofilm Activity of Galla chinensis and Potentially Adjuvant Chemicals on Escherichia coli Biofilm Formation and Preformed Biofilm
Angelina Ning, American Heritage School Boca/Delray

The purpose of this project was to determine whether the effectiveness of G. chinensis treatment on inhibiting E. coli biofilm formation could be reinforced using the naturally occurring chemicals cinnamaldehyde and/or zinc, as well as whether these treatments would be effective in dispersing preformed biofilm.

Mesenchymal Stromal Cells Used for Cellular Cardiomyoplasty As a Way To Treat Myocardial Infarction and Heart Failure
Amarachukwu Okpala, American Heritage School Plantation

Heart failure and heart attacks are two of the most common cardiomyopathies talked about and seen in the United States. Because of this, I was inspired to research if there are any treatments that are being looked at at the moment that show promising results. I was able to come across cellular cardiomyoplasty and mesenchymal stromal cells, leading to the completed research seen in my presentation.

Investigating the Effects of the Growth of Thioflavin T with Omega-6 Acids in the Brainstem in Progressive Supranuclear Palsy
Amay Patel, American Heritage School Boca/Delray

This project was about experimenting different concentrations of omega-6 acids on Thioflavin T in the brainstem in progressive supranuclear palsy.

Strain-Selective in vivo and in silico Structure-Activity Relationship (SAR) of N-Acyl Beta-Lactam Broad-Spectrum Antibiotics
Ayeeshi Poosarla and Ria Kolala, ASDRP

Beta-lactam antibiotics have been used over the years to combat bacterial issues. As bacterial resistance emerges, new antibiotics must be created. By studying the structure-activity relationship between six commercially available antibiotics, targetted synthesis can occur to create novel antibiotics.

Comparison of Nickel Chelator to Current Triple Antibiotic Therapy to Treat Helicobacter pylori Infection
Meena Ramadugu, John F. Kennedy High School

H. pylori is estimated to have infected around 4.4 billion people around the world and has been shown to be the causative agent for acute and chronic gastritis, peptic ulcer, gastric carcinoma and gastric lymphoma. Rapid emergence of antibiotic-resistant strains of H. pylori and poor patient compliance due to side effects have decreased the effectiveness of the current regimen, raising the need for alternative treatments. Last year, a study showed that nickel chelator has an inhibiting effect on H. pylori growth. My current research is comparing the effects of nickel chelator to the current antibiotic regimen.

Sputum-Based mRNA-Targeting Probes in Lung Cancer - A Noninvasive Early Diagnostic Tool
Diya Ramesh, International Academy

Using a multiplex RNA probe in sputum to target the messenger RNAs of known overexpressed tumor marker genes in lung cancer, in order to detect lung cancer early and noninvasively.

Determining the Nattokinase Facilitated Staphyloccocus aureus Biofilm Dispersion Pathway
Raghav Sriram, Carmel High School

Computationally Analyzing Interactions between Nattokinase Binding-Partners via a Novel Stacked Generalization Machine Learning Approach

Genetic Mutation of the SARS-CoV-2 Genome and Its Resulting Infectivity
Amith Vasantha, BASIS Independent Silicon Valley

This project’s purpose was to identify how genetic mutation of the SARS-CoV-2 genome influences its infectivity. SARS-CoV-2 genome sequences were processed using a variant calling pipeline, a Bash shell script incorporating multiple bioinformatics terminal tools like BWA and Samtools. I hypothesized that SARS-CoV-2 spike protein mutations would increase SARS-CoV-2’s infectivity. By analyzing FASTA files for recent, more infectious SARS-CoV-2 lineages, I determined that the SNPs D614G and N501Y repeatedly appeared in these lineages compared to 2020 FASTA files.

Circadian Rhythm, Energy Metabolism, and the Gut Microbiome
Bailey Vergara, American Heritage School Plantation

This review discussed the interrelationship between the circadian rhythm of the host and of the intestinal microbiome and the implications of this new field on energy metabolism, specifically implications relating to obesity. This interrelationship is mainly discussed through possible effects of circadian disruption and through an analysis of different methods by which the circadian rhythm of the host and the rhythm of the microbiome interact.

Shining a Light on Acne
Rhea Zahir, Flint Hill School

My presentation talks about finding the most efficient and cost- effective way to mitigate acne growth. I used a combination of Curcumin and Blue Light to achieve the largest reduction in bacterial growth over a four-day interval.

Undergraduate

Investigating the role of ZF30C in the transcriptional repression of Drosophila engrailed gene
Jessica Escobar, Presbyterian College

My research project involves examining the role of a Zinc finger protein in the process of silencing of gene expression. Gene expression refers to the process of making proteins from DNA and it is a tightly regulated process. Any aberration in this process during development can lead to severe defects or cancer. I will be presenting data from my research where I use the common fruit fly, Drosophila melanogaster, to examine the role of a specific protein in the regulation of a protein that is important in development. Humans share many genes with Drosophila and therefore, we expect to better understand regulation of gene expression from the results of my research.

A Remedy for ur Ailing Soils: Two Variovorax strains with potential for Sustainable Plant Growth Promotion
Akshaya Jayakarunakaran, University of California, Los Angeles

My research project investigates two novel variovorax bacterial strains on their ability to enhance plant growth. The aim is to utilize these bacteria to act as a replacement for chemical growth enhancers.

ADAD: Alcohol Dependence on Alzheimer's Disease
Julie Lee, University of North Carolina at Chapel Hill

If AD pathology can develop long before symptoms, and moderate drinking increases its magnitude, then will higher levels of alcohol drinking when younger also affect AD onset and progression?

CRISPR-Cas Systems in Yersinia Species
Chun-kit Shum, University of Delaware

An analyzation of CRISPR-Cas systems in various Yersinia species through the use of bioinformatics.

Graduate

Mapping Uncharted Biological Territory — Discovering Unique Factors for understanding chlamydial basic biology
Srishti Baid, University of Kansas

My presentation talks very briefly about the identification of a putative transcription factor in Chlamydia trachomatis

Legionnaries' Disease with Abnormal Brain Magnetic Resonance Imaging
Oladoyin Jolaoye, West Virginia School of Osteopathic Medicine

The abstract I will be presenting is about an interesting case of a patient with legionnaires’ disease found on an MRI, this is quite rare in the medical world.

A new mosquito midgut gene AGAP008138 facilitates Plasmodium falciparum transmission in mosquitoes
Julian Ramelow, Florida International University

This project revealed that a unique protein to the Anopheles mosquito (referred to as AGAP008138) facilitates the Malaria parasite into the mosquito midgut cells. Our initial results from the anti-AGAP008138 IgG antibody blocking assay demonstrated that AGAP008138 plays a critical role in malaria transmission through direct interactions with the parasite. Significance: This project revealed a possible new antigen for anti-malarial vaccine development in future by discovering a new protein-protein interaction mechanism.

Distribution and Prevalence of Naegleria fowleri in Freshwater from Alabama Lakes and Ponds
Leigha Stahl, University of Alabama

In this presentation, I will discuss my research on determining the presence of Naegleria fowleri, the brain-eating amoeba, in lakes and ponds of Alabama.

MICROBIOLOGY AND MOLECULAR BIOLOGY

High School

Undergraduate

Graduate

About

Programs

Chapters

Meetings & Events

Members

News

Publications