Posts Tagged computational biology

CellWorks & BioImagene – Successfully bringing IT innovation to Life Sciences

With technology advancement and ability for wider application, interesting things are happening on the boundaries of disciplines, rather than within a given discipline.  Internet and Sofware technology advancement along with fresh “can do” attitude, has infused new energy into Biotech.  CellWorks and BioImagene are two examples, where software stalwarts and serial entrepreneurs like Pradeep Fernandes and Mohan Uttarwar brought innovation to be applied to the healthcare industry.

At event on August 5, 2013, Fernandes and Uttarwar discussed what motivated them to gravitate towards life sciences, after a career in IT and software, the opportunities and challenges in making this transition, and the lessons learned in the process.  They both said, their ignorance about regulatory challenges was a bliss because had they been previously aware of the regulatory hurdles, they might have doubted the whole enterprise.  Both of them encountered a problem in healthcare and realized there was a hole in the market, in addressing the problem.

Fernandes’s wife got thyroid cancer and while making the rounds of the hospitals and doctors, they came to realize that despite existence of huge amount of big data and large quantities of freely available published data, few people were connecting the dots for efficient diagnosis and treatment. Serendipitous discussions with his good friend and colleague (an engineer) and his wife, a PhD in biology, led them to investigate whether modeling can be used to connect the dots and predict biological effects computationally, bridging two disciplines. R&D costs in big pharma are skyrocketing and it costs about $1to $4 B per drug, before it is available for commercial use.  During the development stage, 40% of drugs fail due to lack of efficiency (that is, they simply do not work), and other 40% fail due to toxicity (bad side effects), said Fernandes.  In the long chain of drug development, Fernandes focused on Biology.  “If the biology is wrong, no amount of chemistry is going to make it safe or potent”, said Fernandes.  The challenge is that despite technology advancement, drug development process remains unchanged and key decisions regarding validation of drugs happen very late in clinical trials.   Bringing the advancement in computational algorithms and informatics, his company, CellWorks extracts information from large data sets and also predicts information through abstraction modeling and simulation.  With an ability to model how pathways interact to represent disease phenotypes, CellWorks has further refined their approach to design therapies with novel mechanism of actions.  Greatly reduced cost and time for drug development has enabled CellWorks to advance their lead candidates for several indications from early design stage to validated animal efficacy studies in very short time period.  CellWorks closed $10M Series B round, at the end of 2012.

Origins of Mohan Uttarwar’s company BioImagene also lie in his keen notice of a problem in the market place that was not addressed.  One of the close family friends had cancer and in accompanying the individual to the medical facilities, he noticed that the medical establishment used glass slides to analyze, store, and share critical information.  This made it costly and challenging to share tissue samples and other information, critical for timely and effective diagnosis and to discuss treatment options.  BioImagene addressed the need by creating a suite of dynamic, image-based technologies that enable image capture, information management, image analysis and virtual sharing of patients’ tissue samples from glass microscope slides.  Their unique software, iScan, enabled viewing, analysis, and managing of tissue images, using a computer.  Emphasizing the need to adapt and change, Uttarwar said, the pivotal moment in the history of the company was when they made the decision to go into hardware.  They built a unique image viewing input device called the iSlide, and a high performance workstation called Crescendo.  Together their products greatly improved workflow efficiency in image archiving and retrieval, remote viewing, and turnaround time and made them an attractive target and the company was acquired for $100M by Roche, in August, 2010.

Both Uttarwar and Fernandes credited their team and EPPIC for their success.  EPPIC was started with a vision to promote networking, entrepreneurship, and mentoring for life science professionals, by a handful of people of Indian origin and has since vastly expanded.  Uttarwar and Fernandes shared about how they received significant coaching and guidance from prominent EPPIC members and veterans like Nagesh Mhatre, at critical periods.  Additionally, through EPPIC, they connected with Artiman Ventures and received funding as well.  With such a strong community to support and energize, “if I can do it, you can do it”, said Uttarwar.  Mohan Uttarwar is the guest speaker on October, 29 at event, in Sunnyvale, and will be speaking about one of his new ventures.   These are free events and all are welcome.

Please note down following dates (September, 18 and October, 26) for upcoming EPPIC events and register at

Enhanced by Zemanta

, , , , , , , , , , , , , , , , , ,

Leave a comment

Computational Biology Applied to Liver Cells in vitro – High Throughput Screen for Drug Toxicity

Dr. Mike Bowles, previously a founder of Com21 and IBeam Broadcasting (both of which went on to huge IPOs) and currently co-founder of Biomatica, talked about the application of Computational Biology to investigate drug toxicity effects, earlier in drug development process.  It is an understatement to say that drug development is very expensive, often costing billions of dollars and years of research.  The primary challenge is determination of long term drug toxicity side effects.  If we can develop and deploy efficient technologies for early prediction of adverse side effects, then the costs of drug development can be noticeably reduced, said Bowles.


However, the toxicity studies often take place relatively late in the process.  During first year of research, the focus is on identifying and validating target molecules from over 5000 compounds.  Toxicity is not studied until much later in the development process.  Liver damage is one of the worst potential side effects of drugs, taken alone or with other medicines.  “We need a paradigm shift”, said Bowles, to include toxicity studies earlier in the development process.  But animal studies are also time consuming and they leave many uncertainties about human risk potential.  Often by the time the animal data is in, too much is invested and it is costly to cancel the compounds.  So there is an incentive to go on, rather than to eliminate compounds with riskier profiles.


Biomatica addresses this challenge by replacing liver toxicology studies on live rats with machine learned models of liver damage that can be run on rat (or human) liver cells grown in culture.  They have built models using microarray data from hepatocytes to predict animal and human toxicity.  Toxicity does not occur through just one pathway and it is a diffused problem.  But microarray can encompass all the changes going inside a cell, at any point in time.  Microarray data is collected on rat liver or rat hepatocytes grown in cultures or human hepatocytes grown in cultures and used to find earlier the compounds that should be eliminated.  Testing costs earlier on live rats and on microarrays are similar.  But at following stages they start adding up, in case of live rats.  For instance, for 36 rats, in later stage, while they come to about $20,000 with microarry, with live rats they go as high as $113,400.  The early results are indicating very good prediction accuracy, said Bowles.  The talk generated a lot of interest and was followed by Q&A.


, , , , , , , , , , ,

Leave a comment

%d bloggers like this: