Criteria

1. Does your idea solve a real problem? Is the problem "big" enough?
2. Can you explain the technical/scientific details?
3. Is your idea feasible and implementable? (within 10 weeks)
4. Do you have visuals to explain your idea/product?
5. Do you understand the financial incentives for who needs this?
6. Can the technical/scientific details be elaborated on in just a few hours?
7. Can you imagine the final type of product with the problem (i.e. well defined enough to know what the target market actually wants)

• Intersection of Neurobiology and Genomics → transcriptional profile of brains to find new biomarkers for neurodegenerative disorders (Mendelian Randomization on GWAS to find risk factors that link to neurodegenerative disorders?)
• Phenotype → genotype association: looking at different visual pathological samples and creating predictive models that learn the biological significance of the phenotype (NASH by Insitro)
• Intersection of ML and Genomics → looking into optimizing accuracy for degraded RNA samples by building bias models on existing parts of the RNA-seq pipeline
• Looking at metabolic encapsulates to optimize delivery of nucleic acid therapies → ie. use of lipid nanoparticles or exploring different alternatives to this delivery process
• Comparative analysis → looking at WGS samples and finding new ways to identify variants or predict gene expression for these variants (DeepMind)
• Looking at properties of protein-protein interaction networks and using algorithms to detect biologically relevant functional modules → understanding protein interaction networks allows us to more easily synthesize and replicate them (https://mukundh-murthy.medium.com/using-cnns-to-identify-transcription-factor-binding-motifs-in-dna-sequences-34cabfd88d80?source=user_profile---------18----------------------------)
• Uncertainty in bioinformatic analysis (large feature size, estimation with algorithms, suboptimal sampling, inaccurate significance levels) -> optimizing this optimizes the accuracy of analysis (https://pure.tue.nl/ws/files/3870195/763166.pdf)
• Biosensors and Signal Processing → Monitoring Patient Disease States for example (at the electrochemical level) or even for brain stuff
• Helping with rapid testing expansion → can we create a system to aggregate testing data or build a pipeline to effectively help with some bottlenecks for antigen tests (could be applied as infra. for developing countries)
• Simulating genetic pathways to determine the function of enzyme and future effects
• Looking at metagenomics for gut microbiome → informing healthy eating and diet decisions (dataset and starter)