My Two Decades in Academia

 

Contributions

 

I describe my contributions over the past two decades in academia. My work has been always of inter-disciplinary in nature and it has been always about analyzing and visualizing data.  You can look at at my Google Scholar Page for a full list.

 

  1. Integrative genomic approach for cancer patient stratification

For a complex disease such as cancer, there are many different ways to characterize it such as genotyping, gene expression profiling and histological and clinical subtyping. This issue has become prominent given the availability of The Cancer Genome Atlas (TCGA). In the data repositories accompanying TCGA, multiple types of molecular, phenotype and clinical data are available for multiple types of cancers. Thus, there is a need for developing innovative computational algorithms for integrating multiple types of genomic and phenotype data. I have worked extensively with Prof. Kun Huang on this topic. More recently, I have been collaborating with Prof. Ewy Mathe, Prof. James Chen, and Prof. Parag Mallick.

a.     Ding H, Wang C, Huang K, Machiraju R. iGPSe: A visual analytic system for integrative genomic based cancer patient stratification. BMC Bioinformatics 15: 203 (2014). PMCID: PMC4227100

b.     Wang C, Machiraju R, Huang K. Breast Cancer Patient Stratification using a Molecular Regularized Consensus Clustering Method. Methods  2014 Jun 1;67(3):304-12, PMCID: PMC4151565

c.     Wang C, Machiraju R, Huang K. Cancer Patient Integrative Stratification via a Two-step Consensus Clustering of Molecular Expression and Clinical Attributes. 2014 AMIA (American Medical Informatics Association) Summit on Translational Bioinformatics, San Francisco, CA, April 2014 (podium presentation).

d.     Wang C, Pecot T, Zynger D, Shapiro C, Machiraju R, Huang K. Identifying survival associated morphological features of triple negative breast cancer using multiple datasets, Journal of American Medical Informatics Association (JAMIA), 20(4): 680-687, 2013. PMCID: PMC3721170.

 

2.    Visual Analytics of biological data

Biological data is large and analysis is alone not sufficient to glean patterns. My group has conducted many exercises that transform the data, develop visual mappings and then allow for meaningful interpretation. The challenges are many and often require an iterative cycle of analysis and visualization often in a tight closed loop. I have collaborated with Prof. Kun Huang on these efforts.

a.     Ding H, Wang C, Huang K, Machiraju R. GRAPHIE: graph based histology image explorer. BMC Bioinformatics. 2015;16 (Suppl 11):S10:1186. PMCID: PMC4547152

b.     Li Q, Zachmann G, David F, Huang K, Machiraju R. Observing Genomics and Phenotypical Patterns in the Developing Mouse Brain.  IEEE Computer Graphics and Applications, September 2014, pp. 88-97.

c.     Ding H, Wang C, Huang K, Machiraju R. iGPSe: A visual analytic system for integrative genomic based cancer patient stratification. BMC Bioinformatics 15: 203 (2014). PMCID: PMC4227100

d.     Yates A, Webb, A, Chamberlin H, Huang K, Machiraju R. Visualizing Relationships in Multidimensional Data with Glyph SPLOMs and Necessity Graphs. Computer Graphics Forum (Proceedings of Eurovis 2014), Vol. 33, No. 3, June 2014, pp. 301-310.

e.      Park D, Moldovan N, Machiraju R, Pcot T, Robust detection and visualization of cytoskeletal structures in fibrillar scaffolds from 3-dimensional confocal image. In Biological Data Visualization (BioVis), 2013 IEEE Symposium on Biological Data Visualization (BioVis)(pp. 25-32).

f.      Janoos F, Mosaliganti K, Xu X, Machiraju R, Wong S. Robust 3D Reconstruction and Identification of Dendritic Spines from Optical Microscopy Imaging, Journal of Medical Image Analysis, (13)1:167-179, 2009. PMCID: PMC2663851

 

3.    Flow structures with or without machine learning

Fluids present a fascinating canvas to explore. We developed several methods over the years in collaboration with Prof. David Thompson. Earlier we relied on the principles and laws of fluid flow and applied methods of mathematics. Given everything, we increasingly explored the use of statistical machine learning. Below are some manuscripts we authored. Some samples are placed below.

a.     Biswas A, Thompson D, He W, Deng Q, Chen C-M, Shen H-W, Machiraju R, Rangarajan A. An Uncertainty-Driven Approach to Vortex Analysis Using Oracle Consensus and Spatial Proximity.  Proceedings of Pacific Visualization 2015, Hangzhou, China, pp. 223-230.

b.     Zhang L, Deng Q, Machiraju R, Rangarajan A, Thompson D, Walters DK, Shen H-W. Boosting Techniques for Physics-Based Vortex Detection. Computer Graphics, Volume 33, No. 1, February 2014, pp. 282-293.

c.     Soni B, Thompson D, Machiraju R. Visualizing Particle/Flow Structure Interactions in Flows in the Small Bronchial Tubes. IEEE Transactions on Visualization and Computer Graphics, (14)6:1412-27, 2008

d.     Jankun-Kelly M, Jiang M, Thompson D, Machiraju R. Vortex Visualization for Practical Engineering Applications. IEEE Transactions on Visualization and Computer Graphics (Proceedings of  Visualization/Information Visualization 2006), 12(5):957-64, 2006.

e.     Thompson D, Machiraju R, Jiang M, Nair J, Craciun G, Venkata S. Physics-Based Feature Mining for Large Data Exploration. IEEE Computing in Science and Engineering, 4(4):22-30, 2002.

f.      Jiang M, Machiraju R, Thompson D. Detection and Visualization of Vortices. Visualization Handbook, edited by C. Johnson and C. Hansen, Academic Press, 2004, pp. 287-301.

 

 

4.    Bioimage informatics as a quantitative phenotyping tool

Imaging is very effective in characterizing the phenotype at the morphological level. However, quantitatively characterizing tissue and cellular level morphology from microscopic images is a challenging task. Since 2004, my laboratories have been working developing pipelines and scalable algorithms for analyzing and 3D reconstruction from large microscopic image sets. The algorithms and pipelines have been applied to quantitatively measure and model mouse placenta development, breast tumor microenvironment, and angiogenesis. The algorithms have also been included in the Insight Toolkit as funded projects by National Library of Medicine. My consistent collaborator in this space include Prof. Jens Rittscher.

a.     Mosaliganti K, Janoos F, Sharp R, Ridgway R, Machiraju R, Huang K, Wenzel P, de Bruin A, Leone G Saltz J. Detection and Visualization of Surface-Pockets to enable  Phenotyping Studies, Special Issue on Mathematical Methods in Biomedical Image Analysis, IEEE Transactions on Medical Imaging, 26(9):1283-90, 2007.

  1. Mosaliganti K, Cooper L, Sharp R, Machiraju R, Huang K, Leone G. Visualization of Cellular Biology Structures from Optical Microscopy Data. IEEE Transactions on Visualization and Computer Graphics, (14)4:863-76, 2008.

c.     Mosaliganti K, Janoos F, Irfanoglu O, Ridgway R, Machiraju R, Huang K, Saltz J, Leone G, Ostrowski M. Tensor classification of N-point correlation function features for histology tissue segmentation, Medical Image Analysis, 13(1): 156-166, 2009.

d.     Singh S, Janoos F, Pecot T, Caserta E, Rittscher J, Leone G, Huang K, Machiraju R. Non-parametric Population Analysis of Cellular Phenotypes. Accepted to Proceedings of the 14th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Toronto, Canada, Sep 2011.

e.     Rittscher J, Machiraju R, Wong STC (Editors). Microscopic Image Analysis for Life Science Applications, Artech House, 2008.

 

5.    Image analysis workflows for neuroscience

With various collaborators, my group has developed many workflows to process data from diffusion tensor imaging (DTI) and functional magnetic resonance imaging (fMRI) studies. These workflows required a detailed understanding of the physics underlying acquisition and a sophisticated characterization of the inherent signal. We also ensured that the methods have a significant use case rooted in mechanistic and cognitive biology. Our collaborators examined resulting mechanisms arising from ischemic stroke and required the definition of normal cognitive states from the disabled and diseased states. Dr. Istvan Morocz was a great collaborator on some of these projects.

a.     Rink C, Christoforidis G, Khanna S, Peterson L, Patel Y, Khanna S, Abduljalil A, Irfanoglu O, Machiraju R, Bergdall V, Sen C: Tocotrienol Vitamin E Protects Against Pre-clinical Canine Ischemic Stroke by Inducing Arteriogenesis, Nature Journal of Cerebral Blood Flow & Metabolism (2011), June 2011, pp. 113. PMCID: PMC3210346

b.     Janoos F, Machiraju R, Singh S, Morocz I, Spatio-Temporal Models of Mental Processes from fMRI. NeuroImage 2011, Jul 15;57(2):362-77.

c.     Irfanoglu MO, Koay CG, Pajevic S, Machiraju R, Basser PJ.  Diffusion Tensor Field Registration in the Presence of Uncertainty. Medical Image Computing and Computer Assisted Intervention Conference (MICCAI) 2009, London, United Kingdom, pp. 181-189.

d.     Janoos F, Machiraju R, Sammet S, Knopp MV, Warfield SK, Mrocz IA, Measuring effects of latency in brain activity with FMRI. International Symposium on Biological Imaging (ISBI), Rotterdam, The Netherlands, April 2010,1141-1144.

 

6.    Support for genotypical studies

Our group has worked with many groups and contributed towards the completion of informatics and data analysis tasks. These projects are invaluable in providing useful training for students of computer science engaging in biological research. Many of these students have assumed very salient positions given their exposure to ongoing research in clinical and laboratories of system biology. Prof. Gustavo Leone was  constant collaborators.

a.     Liu H, Tang X, Srivastava A, Pcot T, Daniel P, Hemmelgarn B, Reyes S, Fackler N, Bajwa A, Kladney R, Koivisto C, Chen Z, Wang Q, Huang K, Machiraju R, Senz-Robles M, Cantalupo M, Pipas J, and Leone G. Redeployment of Myc and E2f13 drives Rb-deficient cell cycles. Nat Cell Biol. 2015 Aug;17(8):1036-48.

b.     Ouseph M, Li J, Chen H-Z, Pcot T, Wenzel P, Thompson JC, Comstock G, Chokshi V, Byrne M, Forde B, Chong J-L, Huang K, Machiraju R, de Bruin A, Leone G. Atypical E2F Repressors and Activators Coordinate Placental Development. Developmental Cell, 2012:22(4):849-862. PMCID: PMC3483796

c.     Chen H-Z, Ouseph M, Li J, Pcot T, Chokshi V, Kent L, Bae S, Byrne M, Duran C, Comstock G, Trikha P, Mair M, Senapati S, Martin C, Gandhi S, Wilson N, Liu B, Huang Y-W, Thompson J, Raman S, Singh S, Leone M, Machiraju R, Huang K, Mo X, Fernandez S, Kalaszcynska I, Wolgemuth D, Sicinski P, Huang T, Jin V, and Leone G. Canonical and Atypical E2Fs Regulate the Mammalian Endocycle. Nature cell biology. 2012:14(11):1192-1202. PMCID: PMC3616487

d.     Wenzel PL, Wu L, deBruin A, Chen W-Y, Dureska G, Sites E, Pan T, Sharma A, Huang K, Ridgway R, Mosaliganti K, Sharp R, Machiraju R, Saltz J, Yamamoto H, Cross JC, Robinson ML, Leone G. Rb is critical in a mammalian tissue stem cell population. Genes and Development, 21(1):85-97, 2007. PMCID: PMC1759903

 

7.    Volume Rendering

This body of work was conducted when I first started out in academia. Prof. Torsten Moeller and Prof. Klaus Moeller were close collaborators on methods to design filters. Other significant collaborator was Prof. Robert Lee who introduced me to diffusion and finite-element techniques and we used those for volume rendering of thick tissue.

 

a.     Sharp R, Adams J, Machiraju R, Lee R, Crane R. Physics-Based Subsurface Visualization of Human Tissue. IEEE Transactions on Visualization and Computer Graphics, (13)3:620-9, 2007.

b.     Pfister H, Lorensen W, Bajaj C, Kindlmann G, Schroeder W, Sobierajski-Avila L, Martin K, Machiraju R, Lee J. Transfer Function Bake-Off. IEEE Computer Graphics and Applications, 21(3):16-22, 2001.

c.     Moeller T, Machiraju R, Muller K, Yagel R. Evaluation and Design of Optimal Filters using a Taylor Series Expansion. IEEE Transactions of Visualization and Graphics, 3(2):184-99, 1997.

d.     Mller T, Machiraju R, Mueller K, Yagel R. Classification and Local Error Estimation of Interpolation and Derivative Filters for Volume Rendering. Symposium on Volume Visualization 1996, San Francisco, CA, October 1996, pp. 71-78.

e.      Machiraju R, Yagel R. Reconstruction Error and Control: A Sampling Theory Approach. IEEE Transactions on Visualization and Computer Graphics, 2(3):364-78, 1996.

 

8.    Vision, Graphics, Animation, and Wavelets

 

Yes, I dabbled in many other topics. Prof. Hanspeter Pfister came along for the ride.

 

a.      Kim Y, Machiraju R, Thompson D. Path-based Control of Smoke Simulations. ACM SIGGRAPH /Eurographics Symposium on Computer Animation, September 2006, pp. 33-42.

b.      Lee J, Machiraju R, Pfister H, Moghaddam B. Estimation of 3D Faces and Illumination from Single Photographs Using a Bilinear Illumination Model. Proceedings of the Eurographics Symposium on Rendering 2005, Konstanz, Germany, pp.73-82.

c.      Lee J, Moghaddam B, Machiraju R, Pfister H. Model Reconstruction from Sihouette Images," Graphics Interface 2003, Halifax, Canada, pp. 20-27.

d.      Craciun G, Jiang M, Thompson D, Machiraju R. Spatial Domain Wavelet Design and Implementation for Computational Datasets. IEEE Transactions on Visualization and Computer Graphics, 11(2):149-59, 2005.

e.      Craciun G, Jiang M, Thompson D, Machiraju R. Spatial Domain Wavelet Design and Implementation for Computational Datasets. IEEE Transactions on Visualization and Computer Graphics, 11(2):149-59, 2005.

f.       Machiraju R, Zhu Z, Fry B, Moorhead R. Structure Significant Representation of Computational Field Simulation Datasets. IEEE Transactions of Visualization and Graphics, 4(2):117-29, 1998.