The 5′ untranslated region (5′ UTR) of the messenger RNA plays a crucial role in the translatability and stability of a molecule. Thus, it is an important component in the design of synthetic biological circuits for high and stable expression of intermediate proteins. Several UTR sequences are patented and used frequently in laboratories. We present a novel model UTRGAN, a Generative Adversarial Network (GAN)-based model designed to generate 5′ UTR sequences coupled with an optimization procedure to ensure a target property such as high expression for a target gene sequence or high ribosome load. We rigorously analyze and show that the model can generate sequences that mimic various properties of natural UTR sequences. Then, we show that the optimization procedure yields sequences that are expected to yield 32% higher expression (up to 7-fold) on a set of target genes and 12% higher ribosome load on average on a set of generated 5′ UTRs (up to 90% for the best 5′ UTR), compared to the initially generated UTR sequences. We also demonstrate that when there is a single target gene of interest, the expected expression increases by 55% on average and up to 100% for certain genes (up to 15-fold for the best 5′ UTR).

Peptide therapeutics are robust and promising molecules for treating diverse disease conditions. These molecules can be developed from naturally occurring or mimicking native peptides, through rational design and peptide libraries. We developed a new platform for the rapid screening of the peptide therapeutics for disease targets. In the course of the study, we aimed to employ our platform to screen a new generation of peptide therapeutics candidates against aggregation prone protein targets. Two peptide drug candidates for the protein aggregation prone diseases namely Parkinson s and Alzheimer s diseases were screened. Currently, there are several therapeutic applications that are only effective in masking or slowing down symptom development. Nonetheless, different approaches are developed for inhibiting amyloid aggregation in the secondary nucleation phase, which is critical for amyloid fibril formation. Instead of targeting secondary nucleated protein structures, we tried to inhibit monomeric amyloid units as a novel approach for halting disease-condition. To achieve this, we combined yeast surface display and phage display library platforms. We expressed alpha-synuclein, amyloid beta 40 and amyloid beta 42 on yeast surface, and we selected peptides by using phage display library. After iterative biopanning cycles optimized for yeast cells, several peptides were selected for interaction studies. All of the peptides have been used in vitro characterization methods which are QCM-D measurement, AFM imaging, and ThT assay, and they have yielded promising results in order to block fibrillization or interact with amyloid units as a sensor molecule candidate. Therefore, peptides are good choice for diverse disease-prone molecule inhibition particularly those inhibiting fibrillization. Additionally, these selected peptides can be used as drugs and sensors to detect disease quickly and halt disease progression.