Difference between revisions of "Ganesh Sivaraman"
Jump to navigation
Jump to search
Ganesh Sivaraman
PhD student
| Line 26: | Line 26: | ||
[[File:HybridMOS2_V1.png|500px|center|Semiconductor-Metal-Semiconductor]] | [[File:HybridMOS2_V1.png|500px|center|Semiconductor-Metal-Semiconductor]] | ||
| + | |||
| + | * Machine learning based molecular classifier : [https://papers.nips.cc/paper/4830-learning-invariant-representations-of-molecules-for-atomization-energy-prediction.pdf coulomb matrix] has been developed as a feature to map and predict molecular properties. My aim is to implement a molecular classifier based on sorted coulomb matrix as feature for molecules. Implementation is done with [https://spark.apache.org/docs/2.0.0/index.html Apache Spark] (Python API) and machine learning is performed with the Spark built in [http://spark.apache.org/docs/latest/mllib-guide.html MLlib]. | ||
<span style="font-size:200%"> '''publications''' → [https://www.researchgate.net/profile/Ganesh_Sivaraman here]</span>. | <span style="font-size:200%"> '''publications''' → [https://www.researchgate.net/profile/Ganesh_Sivaraman here]</span>. | ||
Revision as of 11:22, 30 August 2016
PhD student
| Office: | 1.080 |
|---|---|
| Phone: | +49 711 685-67721 |
| Fax: | +49 711 685-63658 |
| Email: | ganesh _at_ icp.uni-stuttgart.de |
| Address: | Ganesh Sivaraman Institute for Computational Physics Universität Stuttgart Allmandring 3 70569 Stuttgart Germany |
Research
My research involves computational modeling of solid state devices / materials for next generation label free DNA sequencing (and proteomics ) on High-performance computers. The device simulations are performed with the framework of Density functional theory (DFT) combined with Non-Equilibrium Greens Function (NEGF) Formalism.
- Nucleobase interactions with lower diamondoids.
- The solid state device simulation involves gold electrodes embedded with diamond caged molecules (i.e. Diamondoids) for tunneling based electric DNA sequencing devices .
- Mutation and methylations detection
- Material modeling of Semiconducting (2H) / metallic (1T) phase in MoS2 monolayer for novel nanoscale bio-sensing application.
- Machine learning based molecular classifier : coulomb matrix has been developed as a feature to map and predict molecular properties. My aim is to implement a molecular classifier based on sorted coulomb matrix as feature for molecules. Implementation is done with Apache Spark (Python API) and machine learning is performed with the Spark built in MLlib.
publications → here.
Master Thesis
"Effect of The Protein Electric Field on The Spectral Tuning Of A Photosynthetic System" (267 KB)
, 2012, CBBC Group, Sapienza University of Rome, Italy.