Biology subjects

Cao, R.

Publications and source records attributed to Cao, R..

82 records · Page 5Linked to original sources

bioRxiv2020

A transferrable and interpretable multiple instance learning model for microsatellite instability prediction based on histopathology images

BackgroundMicrosatellite instability (MSI) is a negative prognostic factor for colorectal cancer (CRC) and can be used as a predictor of success for immunotherapy in pan-cancer. However, current MSI identification methods are not available for all patients. We propose an ensemble multiple instance learning (MIL)-based deep learning model to predict MSI status directly from histopathology images. DesignTwo cohorts of patients were collected, including 429 from The Cancer Genome Atlas (TCGA-COAD) and 785 from a self-collected Asian data set (Asian-CRC). The initial model was developed and validated in TCGA-COAD, and then generalized in Asian-CRC through transfer learning. The pathological signatures extracted from the model are associated with genotypes for model interpretation. ResultsA model called Ensembled Patch Likelihood Aggregation (EPLA) was developed in the TCGA-COAD training set based on two consecutive stages: patch-level prediction and WSI-level prediction. The EPLA model achieved an area-under-the -curve (AUC) of 0.8848 in the TCGA-COAD test set, which outperformed the state-of-the-art approach, and an AUC of 0.8504 in the Asian-CRC after transfer learning. Furthermore, the five pathological imaging signatures identified using the model are associated with genomic and transcriptomic profiles, which makes the MIL model interpretable. Results show that our model recognizes pathological signatures related to mutation burden, DNA repair pathways, and immunity. ConclusionOur MIL-based deep learning model can effectively predict MSI from histopathology images and are transferable to a new patient cohort. The interpretability of our model by association with genomic and transcriptomic biomarkers lays the foundation for prospective clinical research.

bioengineering↗

bioRxiv2019

Computationally guided high-throughput design of self-assembling drug nanoparticles

Nanoformulations are transforming our capacity to effectively deliver and treat a myriad of conditions. However, many nanoformulation approaches still suffer from high production complexity and low drug loading. One potential solution relies on harnessing co-assembly of drugs and small molecular excipients to facilitate nanoparticle formation through solvent exchange without the need for chemical synthesis, generating nanoparticles with up to 95% drug loading. However, there is currently no understanding which of the millions of possible combinations of small molecules can result in the formation of these nanoparticles. Here we report the development of a high-throughput screening platform coupled to machine learning to enable the rapid evaluation of such nanoformulations. Our platform identified 101 novel self-assembling drug nanoparticles from 2.1 million pairings derived from 788 candidate drugs with one of 2686 excipients, spanning treatments for multiple diseases and often harnessing well-known food additives, vitamins, or approved drugs as carrier materials - with potential for accelerated approval and translation. Given their long-term stability and potential for clinical impact, we further characterize novel sorafenib-glycyrrhizin and terbinafine-taurocholic acid nanoparticles ex vivo and in vivo. We anticipate that this platform could accelerate the development of safer and more efficacious nanoformulations with high drug loadings for a wide range of therapeutics.

bioengineering↗

bioRxiv2019

Deep convolutional neural networks for predicting the quality of single protein structural models

Predicting the global quality and local (residual-specific) quality of a single protein structural model is important for protein structure prediction and application. In this work, we developed a deep one-dimensional convolutional neural network (1DCNN) that predicts the absolute local quality of a single protein model as well as two 1DCNNs to predict both local and global quality simultaneously through a novel multi-task learning framework. The networks accept sequential and structural features (i.e. amino acid sequence, agreement of secondary structure and solvent accessibilities, residual disorder properties and Rosetta energies) of a protein model of any size as input to predict its quality, which is different from existing methods using a fixed number of hand-crafted features as input. Our three methods (InteractQA-net, JointQA-net and LocalQA-net) were trained on the structural models of the single-domain protein targets of CASP8, 9, 10 and evaluated on the models of CASP11 and CASP12 targets. The results show that the performance of our deep learning methods is comparable to the state-of-the-art quality assessment methods. Our study also demonstrates that combining local and global quality predictions together improves the global quality prediction accuracy. The source code and executable of our methods are available at: https://github.com/multicom-toolbox/DeepCovQA

bioinformatics↗

bioRxiv2019

Cascaded-CNN: Deep Learning to Predict Protein Backbone Structure from High-Resolution Cryo-EM Density Maps

Cryo-electron microscopy (cryo-EM) has become a leading technology for determining protein structures. Recent advances in this field have allowed for atomic resolution. However, predicting the backbone trace of a protein has remained a challenge on all but the most pristine density maps (< 2.5[A] resolution). Here we introduce a deep learning model that uses a set of cascaded convolutional neural networks (CNNs) to predict C atoms along a proteins backbone structure. The cascaded-CNN (C-CNN) is a novel deep learning architecture comprised of multiple CNNs, each predicting a specific aspect of a proteins structure. This model predicts secondary structure elements (SSEs), backbone structure, and C atoms, combining the results of each to produce a complete prediction map. The cascaded-CNN is a semantic segmentation image classifier and was trained using thousands of simulated density maps. This method is largely automatic and only requires a recommended threshold value for each evaluated protein. A specialized tabu-search path walking algorithm was used to produce an initial backbone trace with C placements. A helix-refinement algorithm made further improvements to the -helix SSEs of the backbone trace. Finally, a novel quality assessment-based combinatorial algorithm was used to effectively map C traces to obtain full-atom protein structures. This method was tested on 50 experimental maps between 2.6[A] and 4.4[A] resolution. It outperformed several state-of-the-art prediction methods including RosettaES, MAINMAST, and a Phenix based method by producing the most complete prediction models, as measured by percentage of found C atoms. This method accurately predicted 88.5% (mean) of the C atoms within 3[A] of a proteins backbone structure surpassing the 66.8% mark achieved by the leading alternate method (Phenix based fully automatic method) on the same set of density maps. The C-CNN also achieved an average RMSD of 1.23[A] for all 50 experimental density maps which is similar to the Phenix based fully automatic method. The source code and demo of this research has been published at https://github.com/DrDongSi/Ca-Backbone-Prediction.

molecular biology↗

bioRxiv2019

Artificial Intelligence for Bioinformatics: Applications in Protein Folding Prediction

AI recently shows great promise in the field of bioinformatics, such as protein structure prediction. The Critical Assessment of protein Structure Prediction (CASP) is a nationwide experiment that takes place biannually, which centered around analyzing the best current systems for predicting protein tertiary structures. In this paper, we research on available AI methods and features, and then explore novel methods based on reinforcement learning. Such method will have profound implications for R&D in bioinformatics and add an additional platform to the management of innovation in biotechnology.

bioinformatics↗

bioRxiv2019

AngularQA: Protein Model Quality Assessment with LSTM Networks

Quality Assessment (QA) plays an important role in protein structure prediction. Traditional protein QA methods suffer from searching databases or comparing with other models for making predictions, which usually fail. We propose a novel protein single-model QA method which is built on a new representation that converts raw atom information into a series of carbon-alpha (C) atoms with side-chain information, defined by their dihedral angles and bond lengths to the prior residue. An LSTM network is used to predict the quality by treating each amino acid as a time-step and consider the final value returned by the LSTM cells. To the best of our knowledge, this is the first time anyone has attempted to use an LSTM model on the QA problem; furthermore, we use a new representation which has not been studied for QA. In addition to angles, we make use of sequence properties like secondary structure at each time-step, without using any database. Our model achieves an overall correlation of 0.651 on the CASP12 testing dataset. Our experiment points out new directions for QA problem and our method could be widely used for protein structure prediction problem. The software is freely available at GitHub: https://github.com/caorenzhi/AngularQA

bioinformatics↗

bioRxiv2019

Protein tertiary structure modeling driven by deep learning and contact distance prediction in CASP13

Prediction of residue-residue distance relationships (e.g. contacts) has become the key direction to advance protein tertiary structure prediction since 2014 CASP11 experiment, while deep learning has revolutionized the technology for contact and distance distribution prediction since its debut in 2012 CASP10 experiment. During 2018 CASP13 experiment, we enhanced our MULTICOM protein structure prediction system with three major components: contact distance prediction based on deep convolutional neural networks, contact distance-driven template-free (ab initio) modeling, and protein model ranking empowered by deep learning and contact prediction, in addition to an update of other components such as template library, sequence database, and alignment tools. Our experiment demonstrates that contact distance prediction and deep learning methods are the key reasons that MULTICOM was ranked 3rd out of all 98 predictors in both template-free and template-based protein structure modeling in CASP13. Deep convolutional neural network can utilize global information in pairwise residue-residue features such as co-evolution scores to substantially improve inter-residue contact distance prediction, which played a decisive role in correctly folding some free modeling and hard template-based modeling targets from scratch. Deep learning also successfully integrated 1D structural features, 2D contact information, and 3D structural quality scores to improve protein model quality assessment, where the contact prediction was demonstrated to consistently enhance ranking of protein models for the first time. The success of MULTICOM system in the CASP13 experiment clearly shows that protein contact distance prediction and model selection driven by powerful deep learning holds the key of solving protein structure prediction problem. However, there are still major challenges in accurately predicting protein contact distance when there are few homologous sequences to generate co-evolutionary signals, folding proteins from noisy contact distances, and ranking models of hard targets.

bioinformatics↗

bioRxiv2018

Analysis of microbial community structure of pit mud for Chinese strong-flavor liquor fermentation using next generation DNA sequencing of full-length 16S rRNA

The pit is the necessary bioreactor for brewing process of Chinese strong-flavor liquor. Pit mud in pits contains a large number of microorganisms and is a complex ecosystem. The analysis of bacterial flora in pit mud is of great significance to understand liquor fermentation mechanisms. To overcome taxonomic limitations of short reads in 16S rRNA variable region sequencing, we used high-throughput DNA sequencing of near full-length 16S rRNA gene to analyze microbial compositions of different types of pit mud that produce different qualities of strong-flavor liquor. The results showed that the main species in pit mud were Pseudomonas extremaustralis 14-3, Pseudomonas veronii, Serratia marcescens WW4, and Clostridium leptum in Ruminiclostridium. The microbial diversity of pit mud with different quality was significantly different. From poor to good quality of pit mud (thus the quality of liquor), the relative abundances of Ruminiclostridium and Syntrophomonas in Firmicutes was increased, and the relative abundance of Olsenella in Actinobacteria also increased, but the relative abundances of Pseudomonas and Serratia in Proteobacteria were decreased. The surprising findings of this study include that the diversity of intermediate level quality of N pit mud was the lowest, and the diversity levels of high quality pit mud G and poor quality pit mud B were similar. Correlation analysis showed that there were high positive correlations (r > 0.8) among different microbial groups in the flora. Based on the analysis of the microbial structures of pit mud in different quality, the good quality pit mud has a higher microbial diversity, but how this higher diversity and differential microbial compositions contribute to better quality of liquor fermentation remains obscure.

molecular biology↗

bioRxiv2018

Characterizing the Effects of Glutathione as an Immunoadjuvant in the Treatment of Tuberculosis

Mycobacterium tuberculosis (M. tb) is the etiological agent that is responsible for causing tuberculosis (TB), which continues to affect millions of people worldwide, with an ever-increasing resistance to antibiotics. We tested the synergistic effects of N-acetyl cysteine (NAC, the precursor molecule for the synthesis of glutathione) and individual first-line antibiotics typically given for the treatment of TB such as Isoniazid (INH), Rifampicin (RIF), Ethambutol (EMB) and Pyrazinamide (PZA) to improve the ability of macrophages to control intracellular M. tb infection. Glutathione (GSH), a pleiotropic antioxidant molecule has been previously shown to display both antimycobacterial and immune-enhancing effects. Our results indicate that there was not only an increase in beneficial immunomodulatory effects, but a greater reduction in the intracellular viability of M. tb when macrophages were treated with the combination of antibiotics (INH/RIF/EMB or PZA) and NAC.

microbiology↗

bioRxiv2018

Combining mathematical and statistical modeling to simulate time course bulk and single cell gene expression data in cancer with CancerInSilico

Bioinformatics techniques to analyze time course bulk and single cell omics data are advancing. The absence of a known ground truth of the dynamics of molecular changes challenges benchmarking their performance on real data. Realistic simulated time-course datasets are essential to assess the performance of time course bioinformatics algorithms. We develop an R/Bioconductor package, CancerInSilico, to simulate bulk and single cell transcriptional data from a known ground truth obtained from mathematical models of cellular systems. This package contains a general R infrastructure for running cell-based models and simulating gene expression data based on the model states. We show how to use this package to simulate a gene expression data set and consequently benchmark analysis methods on this data set with a known ground truth. The package is freely available via Bioconductor: http://bioconductor.org/packages/CancerInSilico/

systems biology↗