November ABC Seminar: Gennady Gorin, PhD – CalTech

November ABC Seminar: Gennady Gorin, PhD – CalTech

Stochastic foundations for single-cell RNA sequencing

Single-cell RNA sequencing, which quantifies cell transcriptomes, has seen widespread adoption, accompanied by a proliferation of analytic methods. However, there has been relatively little systematic investigation of its best practices and their underlying assumptions, leading to challenges and discrepancies in analysis. I motivate a set of generic, principled strategies for modeling the biological and technical stochasticity in sequencing experiments, and use case studies to illustrate their prospects for the discovery and interpretation of biophysical kinetics.

Research links:

 

Dr. Gennady Gorin is a chemical engineer working at the exciting intersection of bioinformatics, stochastic biophysics, and statistics. He completed his Ph.D. with Lior Pachter at the California Institute of Technology, adapting theory from fluorescence transcriptomics to the unique features of single-cell RNA sequencing. Prior, he completed a B.S./B.A. at Rice University and performed transcriptional modeling research in the Golding laboratory at Baylor College of Medicine. Gennady is transitioning to industrial bioinformatics, and excited about the prospects for rigorous, physics-informed methods in method development.

All Welcome! Note this event will take place on Zoom.

Date: Monday November 20, 2023
Time: 4:00-5:00PM

Click here to be added to our mail list.

For further information about this seminar series, contact tarnoldmages@bwh.harvard.edu

Gerber Lab awarded $3.1 Million Five Year NIH-NIGMS R35 Grant “Probabilistic deep learning models and integrated biological experiments for analyzing dynamic and heterogeneous microbiomes”

Gerber Lab awarded $3.1 Million Five Year NIH-NIGMS R35 Grant “Probabilistic deep learning models and integrated biological experiments for analyzing dynamic and heterogeneous microbiomes”

This work will leverage deep learning technologies to advance the microbiome field beyond finding associations in data, to accurately predicting the effects of perturbations on microbiota, elucidating mechanisms through which the microbiota affects the host, and improving bacteriotherapies to enable their success in the clinic. New deep learning models will be developed that address specific challenges for the microbiome, including noisy/small datasets, highly heterogenous human microbiomes, the need for direct interpretability of model outputs, complex multi-modal datasets, and constraints imposed by biological principles. Computational models and biological experiments will be directly coupled through reinforcing cycles of predicting, testing predictions with new experiments, and improving models. An important objective will also be to make computational tools widely available to the research community, through release of quality open-source software.

RePORTER Link

 

2023-24 ABC Seminar Speakers Announced

2023-24 ABC Seminar Speakers Announced

Developed by the Computational Pathology Division at Brigham and Women’s Hospital, the Advanced Biomedical Computation (ABC) Seminar, held monthly during the academic year, showcases innovative research from around the globe by up-and-coming investigators that are developing and applying advanced computational methods to solve biomedical problems.

Click here to be added to our mail list.

Mahmood receives 2023 Young Mentors Award from Harvard Medical School

Mahmood receives 2023 Young Mentors Award from Harvard Medical School

Faisal Mahmood, PhD received the 2023 Young Mentor Award from Harvard Medical School for his outstanding contributions to mentorship.

The Young Mentor Award is one of several Excellence in Mentoring Awards that HMS established to recognize the value of quality mentoring relationships and the impact they have on the professional development and career advancement in medicine, teaching, research and administration. Created in 2005, the Young Mentor Award honors HMS faculty who are still in the early stages of their career but are devoting their time to providing mentoring for others.

View Press Release

Gerber Lab at ICML Workshop on Computational Biology 2023

Gerber Lab at ICML Workshop on Computational Biology 2023

The ICML Workshop on Computational Biology (WCB) highlights how ML approaches can be tailored to making both translational and basic scientific discoveries with biological data, such as genetic sequences, cellular features or protein structures and imaging datasets, among others. It aims to bring together interdisciplinary ML researchers working in areas such as computational genomics; neuroscience; metabolomics; proteomics; bioinformatics; cheminformatics; pathology; radiology; evolutionary biology; population genomics; phenomics; ecology, cancer biology; causality; representation learning and disentanglement to present recent advances and open questions to the machine learning community.

The Gerber Lab had the following two papers accepted:

Gerber GK, Bhattarai SK, Du M, Glickman MS, Bucci V. Discovery of Host-Microbiome Interactions Using Multi-Modal, Sparse, Time-Aware, Bayesian Network-Structured Neural Topic Models. International Conference on Machine Learning Workshop on Computational Biology, 2023.

Uppal G, Urtecho G, Richardson M, Moody T, Wang HH, Gerber GK. MC-SPACE: Microbial communities from spatially associated counts engine. International Conference on Machine Learning Workshop on Computational Biology, 2023.

 

Algorithmic fairness in artificial intelligence for medicine and healthcare: Nature Biomedical Engineering

Algorithmic fairness in artificial intelligence for medicine and healthcare: Nature Biomedical Engineering

In healthcare, the development and deployment of insufficiently fair systems of artificial intelligence (AI) can undermine the delivery of equitable care. Assessments of AI models stratified across subpopulations have revealed inequalities in how patients are diagnosed, treated and billed. In this Perspective, we outline fairness in machine learning through the lens of healthcare, and discuss how algorithmic biases (in data acquisition, genetic variation and intra-observer labelling variability, in particular) arise in clinical workflows and the resulting healthcare disparities. We also review emerging technology for mitigating biases via disentanglement, federated learning and model explainability, and their role in the development of AI-based software as a medical device.

Chen RJ, Wang JJ, Williamson DFK, Chen TY, Lipkova J, Lu MY, Sahai S, Mahmood F. Algorithmic fairness in artificial intelligence for medicine and healthcare. Nat Biomed Eng. 2023 06; 7(6):719-742. PMID: 37380750.

Mahmood Lab develops deep learning model for transforming tissue images: Nature Biomedical Engineering 2022

Mahmood Lab develops deep learning model for transforming tissue images: Nature Biomedical Engineering 2022

Histological artefacts in cryosectioned tissue can hinder rapid diagnostic assessments during surgery. Formalin-fixed and paraffin-embedded (FFPE) tissue provides higher quality slides, but the process for obtaining them is laborious (typically lasting 12–48 h) and hence unsuitable for intra-operative use. Here we report the development and performance of a deep-learning model that improves the quality of cryosectioned whole-slide images by transforming them into the style of whole-slide FFPE tissue within minutes. The model consists of a generative adversarial network incorporating an attention mechanism that rectifies cryosection artefacts and a self-regularization constraint between the cryosectioned and FFPE images for the preservation of clinically relevant features. Transformed FFPE-style images of gliomas and of non-small-cell lung cancers from a dataset independent from that used to train the model improved the rates of accurate tumour subtyping by pathologists.

Ozyoruk, K.B., Can, S., Darbaz, B. et al. A deep-learning model for transforming the style of tissue images from cryosectioned to formalin-fixed and paraffin-embedded. Nat. Biomed. Eng 6, 1407–1419 (2022). https://doi.org/10.1038/s41551-022-00952-9

Mahmood Lab develops self-supervised deep learning algorithm: Nature Biomedical Engineering 2022

Mahmood Lab develops self-supervised deep learning algorithm: Nature Biomedical Engineering 2022

The adoption of digital pathology has enabled the curation of large repositories of gigapixel whole-slide images (WSIs). Computationally identifying WSIs with similar morphologic features within large repositories without requiring supervised training can have significant applications. However, the retrieval speeds of algorithms for searching similar WSIs often scale with the repository size, which limits their clinical and research potential. Here we show that self-supervised deep learning can be leveraged to search for and retrieve WSIs at speeds that are independent of repository size. The algorithm, which we named SISH (for self-supervised image search for histology) and provide as an open-source package, requires only slide-level annotations for training, encodes WSIs into meaningful discrete latent representations and leverages a tree data structure for fast searching followed by an uncertainty-based ranking algorithm for WSI retrieval. We evaluated SISH on multiple tasks (including retrieval tasks based on tissue-patch queries) and on datasets spanning over 22,000 patient cases and 56 disease subtypes. SISH can also be used to aid the diagnosis of rare cancer types for which the number of available WSIs is often insufficient to train supervised deep-learning models.

Chen, C., Lu, M.Y., Williamson, D.F.K. et al. Fast and scalable search of whole-slide images via self-supervised deep learning. Nat. Biomed. Eng 6, 1420–1434 (2022). https://doi.org/10.1038/s41551-022-00929-8

Gerber Lab’s “MDITRE: Scalable and Interpretable Machine Learning for Predicting Host Status from Temporal Microbiome Dynamics” is mSystems Editor’s Pick

Gerber Lab’s “MDITRE: Scalable and Interpretable Machine Learning for Predicting Host Status from Temporal Microbiome Dynamics” is mSystems Editor’s Pick

Longitudinal microbiome data sets are being generated with increasing regularity, and there is broad recognition that these studies are critical for unlocking the mechanisms through which the microbiome impacts human health and disease. However, there is a dearth of computational tools for analyzing microbiome time-series data. To address this gap, we developed an open-source software package, Microbiome Differentiable Interpretable Temporal Rule Engine (MDITRE), which implements a new highly efficient method leveraging deep-learning technologies to derive human-interpretable rules that predict host status from longitudinal microbiome data. Using semi-synthetic and a large compendium of publicly available 16S rRNA amplicon and metagenomics sequencing data sets, we demonstrate that in almost all cases, MDITRE performs on par with or better than popular uninterpretable machine learning methods, and orders-of-magnitude faster than the prior interpretable technique. MDITRE also provides a graphical user interface, which we show through case studies can be used to derive biologically meaningful interpretations linking patterns of microbiome changes over time with host phenotypes. 

Mahmood Lab’s Pan-cancer integrative histology-genomic analysis is featured on cover of Cancer Cell

Mahmood Lab’s Pan-cancer integrative histology-genomic analysis is featured on cover of Cancer Cell

The rapidly emerging field of computational pathology has demonstrated promise in developing objective prognostic models from histology images. However, most prognostic models are either based on histology or genomics alone and do not address how these data sources can be integrated to develop joint image-omic prognostic models. Additionally, identifying explainable morphological and molecular descriptors from these models that govern such prognosis is of interest. We use multimodal deep learning to jointly examine pathology whole-slide images and molecular profile data from 14 cancer types. Our weakly supervised, multimodal deep-learning algorithm is able to fuse these heterogeneous modalities to predict outcomes and discover prognostic features that correlate with poor and favorable outcomes. We present all analyses for morphological and molecular correlates of patient prognosis across the 14 cancer types at both a disease and a patient level in an interactive open-access database to allow for further exploration, biomarker discovery, and feature assessment.

Chen RJ, Lu MY, Williamson DFK, Chen TY, Lipkova J, Noor Z, Shaban M, Shady M, Williams M, Joo B, Mahmood F. Pan-cancer integrative histology-genomic analysis via multimodal deep learning. Cancer Cell. 2022 Aug 8;40(8):865-878.e6. doi: 10.1016/j.ccell.2022.07.004. PMID: 35944502.