Join us!
Students trained in any Engineering discipline, Physics or Mathematics preferably with a penchant for studying non-linear dynamical systems and comfortable in writing code (Python, MATLAB, etc.) can apply.
Previous experience with active matter, agent based modelling, data-science research are most welcome. If you are interested, please write to me stating why you are interested in the project, with your CV attached.
First posted on August 2024
Inferring the motility of sperm cells from movement data
POSTDOCTORAL FELLOWSHIP
Looking for postdoctoral fellows to work on the project.
Funding: through the IPDF scheme at IIT Madras. (Click here for more information)
The candidate has to apply to IIT Madras under the IPDF scheme, with a proposal drafted with our inputs. There will be 2 rounds of selection, first at the dept level and then at the institute level. The candidate if selected will be given a lab space at IIT Madras to work on the problem for 2 years.
Keywords: Computer vision, Collective dynamics, Physics-informed data science.
About the project:
There are two broad objectives:
-
Extracting the motion of sperm cells from movement videos. Detection can be done in a number of ways that include training the state of the art deep neural networks. This is followed by tracking the movement of these cells and quantifying the interactions between neighbouring cells. Alternatively, optical flow methods can also be employed to quantify the overall movement of the cell collective.
-
Developing an inference module to reliably infer the "true" intrinsic motility of sperm cells from the extracted movement data. This module is expected to disentangle the motility of individual cells from the effects of interactions with the neighbouring cells and debris, to classify sperm cells as motile, non-progressive or immotile.
Looking for students with PhD in Engineering, Physics or Biotechnology, with an experience in computer vision, image analysis, and/or biophysics. Interest in non-linear dynamics and collective dynamics is desirable.
Relevant literature:
How flocks flock: a data-driven approach to collective behaviour
POSTDOCTORAL FELLOWSHIP
Looking for postdoctoral fellows to work on the project in collaboration with:
Prof Vishwesha Guttal, Center for Ecological Sciences, IISc Bangalore.
Funding: through the IPDF scheme at IIT Madras. (Click here for more information)
The candidate has to apply to IIT Madras under the IPDF scheme, with a proposal drafted with our inputs. There will be 2 rounds of selection, first at the dept level and then at the institute level. The candidate if selected will be given a lab space at IIT Madras to work on the problem for 2 years.
About the project:
There are two broad objectives:
-
Extracting movement information from videos: Fish are allowed to school in shallow 2D tanks and imaged from the top. They exhibit fascinating schooling behaviour where the schools stochastically transition between several states: well-polarised, milling-state, un-polarised. To understand the nature of the dynamics in these systems, it is important to track the movement of fish and analyse the order parameters characterising the various states. This becomes challenging when the number of fish are large (>500). Extracting the movement information becomes critical to understanding the non-linear dynamics of fish schooling.
-
Characterize the stochastic dynamics in the system: Previous work in smaller fish groups have revealed that the order observed at the level of the school is a result of the mesoscale noise in the system. We do not yet know how this translates to, in the large fish limit, since the density and velocity coupling is non-trivial and how these impact the noise-structure is not known. Understanding how fish swim from data, could potentially unlock a new class of models (SPDEs) previously unknown in the physics literature.
Looking for students with a PhD in Engineering or Physics with an experience in computer vision and image analysis. Knowledge/penchant for non-linear dynamics, collective behaviour or, PDEs/fluid dynamics, is desirable.
Relevant literature:
-
Jitesh Jhawar, Richard G. Morris, U. R. Amith-Kumar, M. Danny Raj, Tim Rogers, Harikrishnan Rajendran and Vishwesha Guttal, “Noise-Induced Schooling of Fish”, Nature Physics, vol 16, no 4, 2020 (Link).
-
Arshed Nabeel, Ashwin Karichannavar, Shuaib Palathingal, Jitesh Jhawar, David B. Brückner, Danny Raj M and Vishwesha Guttal, “Discovering stochastic dynamical equations from biological time series data”, ArXiV, 2024 (Link).
First posted on March 2024
How flocks flock: a data-driven approach to collective behaviour
POSTDOCTORAL FELLOWSHIP
Looking for postdoctoral fellows to work on the project in collaboration with:
Prof Vishwesha Guttal, Center for Ecological Sciences, IISc Bangalore.
Funding: through the IPDF scheme at IIT Madras. (Click here for more information)
The candidate has to apply to IIT Madras under the IPDF scheme, with a proposal drafted with our inputs. There will be 2 rounds of selection, first at the dept level and then at the institute level. The candidate if selected will be given a lab space at IIT Madras to work on the problem for 2 years.
About the project:
There are two broad objectives:
-
Extracting movement information from videos: Fish are allowed to school in shallow 2D tanks and imaged from the top. They exhibit fascinating schooling behaviour where the schools stochastically transition between several states: well-polarised, milling-state, un-polarised. To understand the nature of the dynamics in these systems, it is important to track the movement of fish and analyse the order parameters characterising the various states. This becomes challenging when the number of fish are large (>500). Extracting the movement information becomes critical to understanding the non-linear dynamics of fish schooling.
-
Characterize the stochastic dynamics in the system: Previous work in smaller fish groups have revealed that the order observed at the level of the school is a result of the mesoscale noise in the system. We do not yet know how this translates to, in the large fish limit, since the density and velocity coupling is non-trivial and how these impact the noise-structure is not known. Understanding how fish swim from data, could potentially unlock a new class of models (SPDEs) previously unknown in the physics literature.
Looking for students with a PhD in Engineering or Physics with an experience in computer vision and image analysis. Knowledge/penchant for non-linear dynamics, collective behaviour or, PDEs/fluid dynamics, is desirable.
Relevant literature:
-
Jitesh Jhawar, Richard G. Morris, U. R. Amith-Kumar, M. Danny Raj, Tim Rogers, Harikrishnan Rajendran and Vishwesha Guttal, “Noise-Induced Schooling of Fish”, Nature Physics, vol 16, no 4, 2020 (Link).
-
Arshed Nabeel, Ashwin Karichannavar, Shuaib Palathingal, Jitesh Jhawar, David B. Brückner, Danny Raj M and Vishwesha Guttal, “Discovering stochastic dynamical equations from biological time series data”, ArXiV, 2024 (Link).