Welcome to Abbasi Lab Webpage

The Abbasi Lab develops interpretable models and algorithms based on machine learning principles to understand functions of the brain and related disorders. To discover and model the functions of neurons, the lab studies applications of machine learning in the analysis of large-scale multi-modal datasets including electrophysiology, calcium imaging, electron microscopy, and spatial gene expression profiles. In the next step and to reconcile our understanding of the brain with related disorders, the Abbasi Lab builds computational tools for scientific discovery in large-scale neurological clinical datasets. This involves interpretable analysis and visualization of medical images, multi-omics datasets, and electronic health records. 

We are hiring!

We have openings for one or multiple post-doctorate fellows to work with us at UCSF on exciting applied problems at the intersection of machine learning and computational/clinical neuroscience. The projects involve predictive modeling, biomedical image and signal processing, interpretation inference, neural networks, visualization, etc. Background in imaging and neuroscience is not required but is a plus!

Contact Reza directly via Reza.AbbasiAsl@ucsf.edu to learn more!

News and Events

Dec 2021: Reza is presenting as an invited speaker at the CMStatistics 2021.

Oct 2021: Roozbeh Farhoodi joins the lab as a new postdoctoral scholar. Welcome, Roozbeh!

August 2021: Congrats to Austin Jang and Shiladitya Dutta, two undergraduate students in the lab, for winning the UC Berkeley URAP fellowship for their research projects!

August 2021: New paper published in Frontiers in Big Data: Structural Compression of Convolutional Neural Networks with Applications in Interpretability.

June 2021: New paper on arXiv: Multi-Modal Prototype Learning for Interpretable Multivariable Time Series Classification.

June 2021: Congrats to Austin Jang and Shiladitya Dutta, for winning the URAP summer fellowship!

May 2021: We got awarded the New Frontiers Research Award from the Sandler Program for Breakthrough Biomedical Research.

May 2021: We are hiring a fully-funded postdoc fellow to lead efforts for our new Neuroscape-based project on computational models of multi-modal biosensor data! Reach out if you are interested!

May 2021: Gavin Cui joins the lab as a bioengineering PhD student. Welcome Gavin!

April 2021: Our NIH grant in collaboration with Neuroscape PIs (Ted Zanto and David Ziegler) on “Neural markers of impending task performance” just got funded!

Feb 2021: One abstract accepted at VSS 2021 on our large-scale standardized survey of neural receptive fields in an entire column in mouse V1.

Feb 2021: We are hiring two new postdoc fellows to join us at UCSF and lead our efforts on developing a computational pipeline to analyze spatial gene expression in mice.

Feb 2021: Reza will be giving a talk at the faculty seminar at the Kavli Institute.

Jan 2021: Reza becomes a Weill Neurohub Investigator.

Jan 2021: We got awarded the Weill Neurohub Next Great Ideas Fund! Our proposed project is focused on spatial gene expression analysis in mice.

Nov 2020: Reza will give an invited talk at the 2020 Basic and Clinical Neuroscience.

Oct 2020: One abstract accepted at CSHL NAISys meeting on interpretable models of neurons in visual cortex.

August 2020: One abstract accepted at JSM 2020 on Revealing Spatial Gene Patterns and Interactions in Mouse Brain via Stability-Driven NMF.

May 2020: One abstract accepted at CNS 2020 on compressed models of neurons in V4!

May 2020: Our paper featured on the cover of Neuron!

March 2020: Reza is now a joint faculty at the Department of Bioengineering and Therapeutic Sciences at UCSF.

March 2020: We are now officially affiliated with Neuroscape!

March 2020: The dataset from our recent eNeuro paper is now available on the CRCNS data repository.

March 2020: Here is a nice read from Allen Institute on our recent Neuron paper.

March 2020: Due to the recent Corona Virus outbreak, the lab has moved to fully remote mode! Hope everyone is staying safe and healthy!

March 2020: We are happy to welcome to the lab Oluwaseun Adegbite as rotation PhD student, Luke Chen as an RA, and five UC Berkeley undergraduates through the URAP program: Kevin Lin, Rohan Divate, Austin Jang, Anmol Parande, Felix Ching!

Feb 2020: Reza will give an invited talk at ITA 2020 on interpretation of neural networks with applications in comp neuroscience.

Jan 2020: We will be present at Cosyne 2020 with one invited talk and one accepted abstract on compressed interpretable models of neurons. Reza will give the invited talk at the workshop on “Scrutinizing models of brain function: from in-silico stimulus synthesis to direct brain perturbation”.

Jan 2020: Our paper titled “Systematic Integration of Structural and Functional Data into Multi-Scale Models of Mouse Primary Visual Cortex” has been accepted at Neuron.

Jan 2020: Our paper titled “Superficial bound of the depth limit of 2-photon imaging in mouse brain” appeared on eNeuro.

Jan 2020: We are now affiliated with Bakar Computational Health Sciences Institute at UCSF! Reza will be a faculty affiliate at the institute.

Dec 2019: Welcome to the lab Neel!

Nov 2019: Welcome to the lab Ahyeon and Franklin!

Nov 2019: Reza joined the UC Berkeley-UCSF graduate program in bioengineering. We have multiple openings for rotation students!

Oct 2019: Our paper titled “Definitions, methods, and applications in interpretable machine learning” appeared on PNAS.

Oct 2019: Reza is presenting at SfN on functional imaging of excitatory neurons in an entire column in mouse visual cortex. Here is the abstract.

Sep 2019: Abbasi Lab officially started!

Our lab’s vision is to better understand neuroscience and neuro-medicine by designing specialized interpretable tools from machine learning and statistics.

Functional modeling in brain

We develop models and algorithms based on principals in statistics and machine learning to understand functions in the visual cortex of brain. To achieve this, we study large-scale datasets with a variety of modalities such as electrophysiology, calcium imaging, and electron microscopy, and gene expression.

Specific projects:

• Models based on deep neural network are effective in predicting single neuron responses in primate visual cortex. Despite their high predictive accuracy, these models are generally difficult to interpret. This limits their applicability in characterizing neuron function. We investigate methods and algorithms to elicit interpretation and visualization of models of neurons based on deep neural networks. To achieve this, we develop frameworks based on stability, compression, and sparsity.

• Characterizing the relationship between neural function and connectivity is an eminent question of visual sensory processing. In order to explore this relationship, we are analyzing recorded visual responses from pan-excitatory neurons within an 800X800 um region of primary visual cortex, spanning all visual layers from pia to white matter. This includes 750 2-photon and 35 3-photon calcium imaging planes spaced by ~16 um. Our goal is to examine the single-cell and population activity in primary visual cortex, and along with Electron Microscopic reconstruction from the same tissue, will serve as a valuable resource to study the functional connectome in mouse cortex.

Computational clinical neuroscience

In order to bridge our understanding of the brain to diagnosis, we build tools to study and visualize large-scale neurological clinical datasets. Our goals are to relate patient health records to observed symptoms as well as help clinicians to easily benefit from the information hidden in the data.  This involves analyzing and visualizing medical images and structured and unstructured clinical documents.

Specific projects:

• In collaboration with Neuroimmunology and Glial Biology division at UCSF Neurology, we are developing quantitative methods to analyze multi-modal data from patients with neurological disease. The data includes one-of-a-kind measurements of MRI, genetic variants, electronic health record, etc from thousands of patients. Our goal is to develop analysis and modeling frameworks to identify scientifically meaningful patterns in patient data that are associated with the disease processes.

Interpretable Machine Learning

Guided by scientific questions from neuroscience and biomedicine, we are interested in the general problem of interpreting machine learning models. In the past decade, research in machine learning has been principally focused on the development of algorithms and models with high predictive capabilities. However, interpreting these models remains a challenge, primarily because of the large number of parameters involved. We investigate methods based on statistical principles to build more interpretable machine learning models. 

Reza is an Assistant Professor in the Department of Neurology and the Department of Bioengineering and Therapeutic Sciences at UCSF. He is a core faculty member at the UCSF Neuroscape Center, a Weill Neurohub Investigator, and the Director of Data Analytics and Visualization at the UCSF Weill Institute for Neuroscience.

Before joining UCSF, Reza was a scientist at the Allen Institute for Brain Science in Seattle. He completed his PhD and MSc in Electrical Engineering and Computer Sciences at UC Berkeley in 2018, where he worked with Bin Yu and Jack Gallant to develop interpretable machine learning tools with applications in computational neuroscience. Reza received his MSc in Biomedical Engineering from Sharif University of Technology in 2013 and BSc in Electrical Engineering from Amirkabir University of Technology (Tehran Polytechnic) in 2010. He is the recipient of the New Frontiers Research Award from the Sandler Program for Breakthrough Biomedical Research (PBBR) in 2021, and the Eli Jury Award from UC Berkeley, Department of Electrical Engineering and Computer Sciences in 2018. He received the May J. Koshland Fund in Memory of H.A. Jastro Award from UC Berkeley Graduate Division in 2016, the Excellence Award in Biomedical Engineering from Sharif University of Technology in 2013, and the Excellence Award in Electrical Engineering from Tehran Polytechnic in 2010. 

Roozbeh is a postdoctoral scholar at the Abbasi Lab working at the intersection of computational neuroscience and machine learning. His project is focused on understanding the relationships between neural structure and neural activities in the primary visual cortex. Before joining UCSF, he was a postdoc at the University of Pennsylvania under the supervision of Konrad Kording where he studied the theory of deep learning and how the morphologies of neurons contribute to the computation in the brain. Roozbeh received his Ph.D. in Mathematics from Sharif University in 2017 where he worked on the emergent patterns in neural networks.

Gavin is a PhD student in the UC Berkeley-UCSF joint Bioengineering program. Before coming to the program, Gavin was an undergraduate researcher at McGill University, where he generated MRI-based simulations of transcranial Direct Current Stimulation. Gavin’s background in Neuroscience and Medical Imaging prompts him to explore projects that apply machine learning methods for predicting neurological disorder prognosis. Currently, Gavin’s focus in the lab is to build models that can combine longitudinal MRI data and genomics data to improve Multiple sclerosis prognosis.

Neel is a physician-scientist who is completing clinical training in Neurology at UCSF. He has a background in both biology and mathematics which has him to pursue practical medical questions with rigorous, data-driven approaches. He has joined the lab to combine quantitative approaches with medicine and biology to improve diagnosis and prognosis of disorders and diseases and develop predictive models for disorders and diseases that lead to evidence-based interventions (related to the idea of “precision medicine”). Neel works on basic science questions with the Kriegstein lab.

Gaurav is an undergraduate Electrical Engineering and Computer Science major at UC Berkeley. Gaurav is currently working on designing an interpretable machine learning framework for multi-variable time-series classification. In the future, Gaurav will apply his framework to multi-modal bio-sensing datasets, as well as continue to develop interpretable machine learning methods. Prior to joining the lab, he worked on research in applying reinforcement learning for network scheduling.

Austin is an undergraduate student studying Electrical Engineering and Computer Science at UC Berkeley. At Abbasi Lab, he is interested in developing tools to interpret and train deep learning models to understand neurological responses to image stimuli. His project involves assessing biophysical models of the brain through state-of-the-art machine learning models.

Before joining the lab, Ahyeon was an undergraduate researcher at the Redwood Center for Theoretical Neuroscience, where she conducted work at the intersection of machine learning and neuroscience. After working briefly at the Computational Neurobiology Laboratory at the Salk Institute, she worked as a research associate at LBNL where she used statistical machine learning methods on medical record data to predict traumatic brain injury outcomes. Ahyeon is interested in developing more interpretable deep learning models to understand the mechanism of neurological disorders and improve clinical diagnoses through precision medicine tools. 

Zeyu is an undergraduate student majoring in Computer Science and Math at UC Berkeley. At Abbasi Lab, he is working on designing a method to estimate the stability of patterns extracted using deep autoencoders. His project involves leveraging these stability-driven analyses to identify low-dimensional representations in spatial gene expression datasets.

Undergraduate researchers

Rohan Divate, UC Berkeley

Austin Jang, UC Berkeley

Arbaaz Muslim, UC Berkeley

Kevin Chen, UC Berkeley

Shiladitya Dutta, UC Berkeley

Former members

Franklin Heng, Research Assistant

Oluwaseun Adegbite, Rotation PhD student

Anmol Parande, Undergraduate researcher

Join us!

We have several openings for post-docs, graduate and undergraduate students and software developers. Contact Reza directly at Reza.AbbasiAsl@ucsf.edu to learn more!

Pre-prints

Gaurav Ghosal, R. Abbasi-Asl. Multi-Modal Prototype Learning for Interpretable Multivariable Time Series Classification, 2021.

R. Abbasi-Asl, A. Ghaffari and E. Fatemizadeh. Robust Image Registration via Empirical Mode Decomposition, arXiv preprint, 2020. (WEB)

R. Abbasi-Asl∗, Y. Chen∗, A. Bloniarz, M. Oliver, Ben Willmore, J. L. Gallant and B. Yu. The DeepTune framework for modeling and characterizing neurons in visual cortex area V4, bioRxiv preprint, 2019. (WEB)

Peer-Reviewed Publications

R. Abbasi-Asl and B. Yu. Structural Compression of Convolutional Neural Networks with Applications in Interpretability, Frontiers in Big Data, 2021. (WEB)

Y.N. Billeh, B. Cai, S.L. Gratiy, K. Dai, R. Iyer, N.W. Gouwens, R. Abbasi-Asl, X. Jia, J.H. Siegle, S.R. Olsen, C. Koch, S. Mihalas, and A. Arkhipov. Systematic Integration of Structural and Functional Data into Multi-Scale Models of Mouse Primary Visual Cortex, Neuron, 2020. (Featured cover Image, WEB, bioRxiv version, DATA, Allen Institute Coverage)

K. Takasaki, R. Abbasi-Asl, J. Waters. Superficial bound of the depth limit of 2-photon imaging in mouse brain, eNeuro, 2020. (WEB, bioRxiv version , DATA)

J. Murdoch∗, C. Singh∗, K. Kumbier†, R. Abbasi-Asl†, and B. Yu. Interpretable machine learning: definitions, methods, and applications, PNAS, 2019. (WEB, NEWS)

R. Abbasi-Asl, M. Keshavarzi, D. Y. Chan. Brain-Computer Interface in Virtual Reality, 9th International IEEE EMBS Conference on Neural Engineering (NER), 2019.

K. Takasaki, J. Larkin, R. Abbasi-Asl, D. Denman, D. Millman, S. de Vries, M. Takeno, N. M da Costa, R C. Reid, J. Waters. 3-Photon Calcium Imaging of Deep Cortical Layers for Functional Connectomics, Optics and the Brain, 2019.

A. Arkhipov, N. W. Gouwens, Y. N. Billeh, S. Gratiy, R. Iyer, Z. Wei, Z. Xu, R. Abbasi-Asl, J. Berg, M. Buice, N. Cain, N. da Costa, S. de Vries, D. Denman, S. Durand, D. Feng, T. Jarsky, J. Lecoq, B. Lee, L. Li, S. Mihalas, G. K. Ocker, S. R. Olsen, R. C. Reid, G. Soler-Llavina, S. A. Sorensen, Q. Wang, J. Waters, M. Scanziani, C. Koch. Visual physiology of the Layer 4 cortical circuit in silico. PLOS Comp. Bio., 2018. (WEB, DATA)

R. Abbasi-Asl and B. Yu. Interpreting Convolutional Neural Networks Through Compression, Neural Information Processing Conference 2017 Symposium on Interpretable Machine Learning (NIPS IML Symp.), Long Beach, CA, 2017. (WEB)

R. Abbasi-Asl, C. Pehlevan, B. Yu and D. B. Chklovskii. Do retinal ganglion cells project natural scenes to their principal subspace and whiten them? IEEE Proceedings of 50th Asilomar Conference on Signals, Systems and Computers (ACSSC), Pacific Grove, CA, 2016.

A. Ahadi, H. Hayati, J. Mitra, R. Abbasi-Asl, K. Awodele. A new method for estimating the longevity and degradation of photovoltaic systems considering weather states. Frontiers in Energy. 2016; 10 (3) :277 – 285.

R. Abbasi-Asl, R. Khorsandi, B. Vosooghi Vahdat. Hammerstein-Wiener Model: A New Ap- proach to the Estimation of Formal Neural Information. Basic and Clinical Neuroscience. 2012; 3 (4) :45-51. (WEB) (A conference paper version published in IEEE Proceedings of The 16th CSI Symposium on Artificial Intelligence and Signal Processing (AISP))

A. Ghanbari, R. Abbasi-Asl, A. Ghaffari, E. Fatemizadeh. Automatic B-spline Image Registra- tion Using Histogram-based Landmark Extraction. In: IEEE-EMBS Conference on Biomedical Engineering & Sciences (IECBES), 2012. (WEB)

R. Abbasi-Asl, R. Khorsandi, Sh. Farzampour, E. Zahedi. Estimation of Muscle Force with EMG Signals Using Hammerstein-Wiener Model. In: IFMBE Proceedings 5th Kuala lumpur International Conference on Biomedical Engineering (BIOMED), Malaysia, Vol. 35, pp. 157- 160, 2011. (WEB)

R. Abbasi-Asl, E. Fatemizadeh. MMRO: A Feature Selection Criterion for MR Images Based on Alpha Stable Filter Responses. In: IEEE Proceedings of 7th Iranian Conference on Machine Vision and Image Processing (MVIP), Iran, 2011.

S. Abadpour, R. Abbasi-Asl, Gh. Moradi. Analysis of Push-Push Oscillators and Designing a Push-Push Oscillator in S Band. In: IEEE Proceedings of 10th Mediterranean Microwave Symposium (MMS), Turkey, 2010.

Selected Conference Abstracts

R. Abbasi-Asl, Arbaaz Muslim, J. Larkin, K. Takasaki, D. Millman, D. Denman, J. Lecoq, A.Arkhipov, N. W. Gouwens, J. Waters, R. C. Reid, S. E. J. DE Vries. A large-scale standardized survey of neural receptive fields in an entire column in mouse V1. VSS, 2021

R. Abbasi-Asl and B. Yu. Interpretable modeling of neurons in cortical area V4 via compressed convolutional neural networks. CNS, 2020.

R. Abbasi-Asl and B. Yu. Compressed interpretable models of neurons in cortical area V4, Cosyne, 2020.

R. Abbasi-Asl, J. Larkin, K. Takasaki, D. Millman, D. Denman, J. Lecoq, A. Arkhipov, N. W. Gouwens, J. Waters, R. C. Reid, S. E. J. DE Vries. Functional imaging of excitatory neurons in an entire column in mouse visual cortex, SfN, 2019.

G. K. Ocker, D. Millman, P. Ledochowitsch, M. D. Oliver, R. Abbasi-Asl, M. A. Buice, S. E. J. DE Vries. Measuring the effects of locomotion on visual activity in the mouse cortex in the Allen Brain Observatory, SfN, 2019.

Y.N. Billeh, B. Cai, S.L. Gratiy, K. Dai, R. Iyer, R. Abbasi-Asl, X. Jia, J.H. Siegle, S.R. Olsen, C. Koch, S. Mihalas, and A. Arkhipov. Systematic integration of structural and functional data into a multi-scale model of mouse primary visual cortex, SfN, 2019.

Y.N. Billeh, B. Cai, S.L. Gratiy, K. Dai, R. Iyer, N.W. Gouwens, R. Abbasi-Asl, X. Jia, J.H. Siegle, S.R. Olsen, C. Koch, S. Mihalas, and A. Arkhipov. Systematic Integration of Experimental Data in Biologically Realistic Models of Mouse Primary Visual Cortex: Insights and Predictions, CNS, 2019.

R. Abbasi-Asl, M. Keeshavarzi, D. Y. Chan. Brain-Computer Interface in Virtual Reality, 7th International BCI Meeting, Pacific Grove, CA, 2018. (Accepted abstract and poster presentation)

R. Abbasi-Asl, Y. Chen, A. Bloniarz, M. Oliver, J. L. Gallant and B. Yu. Deep Nets Meet Real Neurons: Pattern Selectivity in V4, 8th International Workshop on Statistical Analysis of Neuronal Data (SAND8), Pittsburgh, PA, 2017. (Accepted abstract and poster presentation)

R. Abbasi-Asl, Y. Chen, A. Bloniarz, J. L. Gallant and B. Yu. Artificial Neurons Meet Real Neurons: Pattern Selectivity In V4, INFORMS Annual Meeting, Nashville, TN, 2016. (Invited oral presentation)

R. Abbasi-Asl, Y. Chen, A. Bloniarz, J. Mairal, J. L. Gallant and B. Yu. Explaining V4 Neurons Pattern Selectivity via Convolutional Neural Network, 7th International Workshop on Statistical Analysis of Neuronal Data (SAND7), Pittsburgh, PA, 2015. (Accepted abstract and poster presentation)

R. Abbasi-Asl, R. Khorsandi, N. Mozaffari, Sh. Farzampour. A Novel Model for Estimation of the Forearm Motor Forces Using EMG Signals. In: Proceedings of The 1st International Congress of Neuromuscular and Electrodiagnostic Medicine, Iran, 2011. (Accepted abstract and poster presentation)

Address:
Joan and Sanford I. Weill Neurosciences Building
1651 4^th St, Room 611L (Reza’s office: 612B)
San Francisco, CA 94143
(Zipcode for FedEx shipments: 94158)

E-mail:
Reza.AbbasiAsl@ucsf.edu

https://www.ucsf.edu/cgr/cgr-projects/mission-bay-neurosciences-research-building

Tweets by Reza_Abbassi