Venkata Soumith Jonnakuti, a fifth-year MD/Ph.D. candidate at Baylor College of Medicine, under the mentorship of Drs. Zhandong Liu, Mirjana Maletić-Savatić, and Hari Yalamanchili, faculty at the Jan and Dan Duncan Neurological Research Institute (Duncan NRI) at Texas Children’s Hospital and Baylor College, has developed an innovative computational tool called 'Poly(A)Miner-Bulk tool' and used it to study the underlying molecular mechanism for X-Linked Dystonia Parkinsonism (XDP), a hereditary brain disease that singularly affects people from the Island of Panay in the Philippines. Similar to common movement disorders like Parkinson’s disease and dystonia, XDP involves a progressive loss of neurons in brain regions that control movement, resulting in significant disability (e.g., problems in walking, talking, and swallowing) and decreased lifespan. There is currently no cure for XDP and there is an urgent need to develop new therapeutic options.
For this discovery, Jonnakuti recently won third prize in the Therapeutics and Research Solutions Coding Challenge for X-Linked Dystonia Parkinsonism (XDP), a competition hosted by the Massachusetts General Hospital (MGH), The Collaborative Center for X-Linked Dystonia Parkinsonism, and the Sunshine Care Foundation to increase awareness about XDP and to provide a global platform for scientists, clinicians, data scientists, and community members to brainstorm and find innovative solutions for this rare disorder.
This project was initiated by Jonnakuti's observation that the core protein machinery involved in a fundamental molecular process called alternative polyadenylation (APA) are differentially expressed in XDP neural stem cells. This was the first clue that alternative polyadenylation (APA) was involved in the pathogenesis of this rare disorder.
The central dogma of molecular biology is that DNA (‘the genetic blueprint’) is converted to messenger RNA (mRNA), which undergoes chemical modifications (post-transcriptional modifications) that regulate when, where, and how much of its protein product should be produced. The addition of a long sequence of multiple adenine residues (the so-called poly A tail) to precursor mRNA is one such post-transcriptional maturation step that is important for the stability and the localization of mRNAs as well as for their efficient conversion to proteins. Alternative polyadenylation (APA) is a phenomenon in which mRNA molecules with varying poly A tails originate from distinct polyadenylation sites of a single gene and there is increasing understanding of its role in regulating gene expression, particularly in the context of neurodegenerative disorders.
“I collaborated with the Bragg Lab at MGH who are experts in XDP biology to better understand the underlying changes in the molecular process of APA using a computational science approach and to identify how those changes lead to XDP,” said Venkata Jonnakuti, who is a student in the Medical Scientist Training Program at Baylor.
“To test if this process is involved in XDP pathogenesis, Venkata used the tool, Poly(A)Miner-Bulk, that he developed,” Dr. Hari Yalamanchili, also an assistant professor at Baylor College, said. “PolyA-miner-Bulk is equipped with a deep learning module that can recover APA dynamics that are often masked by sequence noise, by taking advantage of the knowledge of the underlying sequence features (motifs) and biochemical mechanisms in sequencing protocols.”
“This approach is a huge leap in inferring polyadenylation sites from bulk RNA-Seq data. In addition, Venkata used vector projections to accurately infer intermediate APA site dynamics that are often missed using other current approaches,” added Dr. Zhandong Liu, who is also the Chief of Computation Sciences at Texas Children’s and an associate professor at Baylor.
“Notably, using this tool on a subset of the Religious Orders Study/Memory and Aging Project (ROSMAP) dataset, Venkata identified altered APA dynamics in Alzheimer's samples,” Dr. Mirjana Maletic-Savatic, who is also the director of the Human Disease Cellular Model Core at Baylor’s Intellectual and Developmental Disabilities Center and an associate professor at Baylor, said. “Interestingly, a few APA genes are being actively investigated as potential therapeutic targets for Alzheimer’s disease. These preliminary findings support the feasibility of using our newly-developed tool to identify novel therapeutic targets and also validate our strategy of looking for changes in alternative polyadenylation and other post-transcriptional processes (e.g. histone modification).”
The Duncan NRI team plan to use the Poly(A)Miner-Bulk tool to create a signature APA pattern for XDP, which can then be potentially used to develop specific and effective diagnostic and treatment options for XDP patients. They are hopeful that if successful, this strategy can later also be used to develop diagnosis and therapies for Parkinson’s disease and different forms of dystonia.
This research was supported by a fellowship from the Gulf Coast Consortia on the NLM Training Program in Biomedical Informatics and Data Science T15 LM0070943.