Badri Adhikari

Main Research Interests

• Deep learning and Bioinformatics
• Protein inter-residue distance prediction
• Deep learning for improving human health

Selected Research Involvements

• Development of various tools for modeling 3D structures of proteins and chromosomes, supported by NSF and NIH.
• Development of deep learning-based tools and methods for solving various problems in the field of protein structure prediction, supported by NSF, NIH, NVIDIA, and Google.
• Artificial intelligence methods Autonomous Environmental Monitoring and Management, supported by NASA.
• Machine learning methods development for collaborators in various departments at UMSL.

• $163,535, National Science Foundation CISE CRII, 2020 (PI).

UM/UMSL Grants/Contracts

• $6,450, UMSL Research Award, 2019 (PI).
• $5,415, UMSL Research Award, 2018 (PI).

Select publications

• Adhikari, B. A fully open-source framework for deep learning protein real-valued distances. Nature Scientific Reports10,2045–2322 (2020).
• Adhikari, B. DEEPCON: protein contact prediction using dilated convolutional neural networks with dropout. Bioinformatics 36, 470–477 (2020).
• Adhikari, B., Hou, J. & Cheng, J. DNCON2: improved protein contact prediction using two-level deep convolutional neural networks. Bioinformatics 34, 1466–1472 (2018).
• Adhikari, B., Trieu, T. & Cheng, J. Chromosome3D: reconstructing three-dimensional chromosomal structures from Hi-C interaction frequency data using distance geometry simulated annealing. BMC genomics 17, 886 (2016).
• Adhikari, B., Bhattacharya, D., Cao, R. & Cheng, J. CONFOLD: residue-residue contact-guided ab initio protein folding. Proteins: Structure, Function, and Bioinformatics 83, 1436–1449 (2015).