An integrative framework for clinical diagnosis and knowledge discovery from exome sequencing data.

Non-silent single nucleotide genetic variants, like nonsense changes and insertion-deletion variants, that affect protein function and length substantially are prevalent and are frequently misclassified. The low sensitivity and specificity of existing variant effect predictors for nonsense and indel variations restrict their use in clinical applications. We propose the Pathogenic Mutation Prediction (PMPred) method to predict the pathogenicity of single nucleotide variations, which impair protein function by prematurely terminating a protein's elongation during its synthesis. The prediction starts by monitoring functional effects (Gene Ontology annotation changes) of the change in sequence, using an existing ensemble machine learning model (UniGOPred). This, in turn, reveals the mutations that significantly deviate functionally from the wild-type sequence. We have identified novel harmful mutations in patient data and present them as motivating case studies. We also show that our method has increased sensitivity and specificity compared to state-of-the-art, especially in single nucleotide variations that produce large functional changes in the final protein. As further validation, we have done a comparative docking study on such a variation that is misclassified by existing methods and, using the altered binding affinities, show how PMPred can correctly predict the pathogenicity when other tools miss it. PMPred is freely accessible as a web service at https://pmpred.kansil.org/, and the related code is available at https://github.com/kansil/PMPred.

Transfer learning for drug-target interaction prediction.

MOTIVATION: Utilizing AI-driven approaches for drug-target interaction (DTI) prediction require large volumes of training data which are not available for the majority of target proteins. In this study, we investigate the use of deep transfer learning for the prediction of interactions between drug candidate compounds and understudied target proteins with scarce training data. The idea here is to first train a deep neural network classifier with a generalized source training dataset of large size and then to reuse this pre-trained neural network as an initial configuration for re-training/fine-tuning purposes with a small-sized specialized target training dataset. To explore this idea, we selected six protein families that have critical importance in biomedicine: kinases, G-protein-coupled receptors (GPCRs), ion channels, nuclear receptors, proteases, and transporters. In two independent experiments, the protein families of transporters and nuclear receptors were individually set as the target datasets, while the remaining five families were used as the source datasets. Several size-based target family training datasets were formed in a controlled manner to assess the benefit provided by the transfer learning approach. RESULTS: Here, we present a systematic evaluation of our approach by pre-training a feed-forward neural network with source training datasets and applying different modes of transfer learning from the pre-trained source network to a target dataset. The performance of deep transfer learning is evaluated and compared with that of training the same deep neural network from scratch. We found that when the training dataset contains fewer than 100 compounds, transfer learning outperforms the conventional strategy of training the system from scratch, suggesting that transfer learning is advantageous for predicting binders to under-studied targets. AVAILABILITY AND IMPLEMENTATION: The source code and datasets are available at https://github.com/cansyl/TransferLearning4DTI. Our web-based service containing the ready-to-use pre-trained models is accessible at https://tl4dti.kansil.org.

Loss of the Nuclear Envelope Protein LAP1B Disrupts the Myogenic Differentiation of Patient-Derived Fibroblasts.

Lamina-associated polypeptide 1 (LAP1) is a ubiquitously expressed inner nuclear membrane protein encoded by TOR1AIP1, and presents as two isoforms in humans, LAP1B and LAP1C. While loss of both isoforms results in a multisystemic progeroid-like syndrome, specific loss of LAP1B causes muscular dystrophy and cardiomyopathy, suggesting that LAP1B has a critical role in striated muscle. To gain more insight into the molecular pathophysiology underlying muscular dystrophy caused by LAP1B, we established a patient-derived fibroblast line that was transdifferentiated into myogenic cells using inducible MyoD expression. Compared to the controls, we observed strongly reduced myogenic differentiation and fusion potentials. Similar defects were observed in the C2C12 murine myoblasts carrying loss-of-function LAP1A/B mutations. Using RNA sequencing, we found that, despite MyoD overexpression and efficient cell cycle exit, transcriptional reprogramming of the LAP1B-deficient cells into the myogenic lineage is impaired with delayed activation of MYOG and muscle-specific genes. Gene set enrichment analyses suggested dysregulations of protein metabolism, extracellular matrix, and chromosome organization. Finally, we found that the LAP1B-deficient cells exhibit nuclear deformations, such as an increased number of micronuclei and altered morphometric parameters. This study uncovers the phenotypic and transcriptomic changes occurring during myoconversion of patient-derived LAP1B-deficient fibroblasts and provides a useful resource to gain insights into the mechanisms implicated in LAP1B-associated nuclear envelopathies.

Data Centric Molecular Analysis and Evaluation of Hepatocellular Carcinoma Therapeutics Using Machine Intelligence-Based Tools.

PURPOSE: Computational approaches have been used at different stages of drug development with the purpose of decreasing the time and cost of conventional experimental procedures. Lately, techniques mainly developed and applied in the field of artificial intelligence (AI), have been transferred to different application domains such as biomedicine. METHODS: In this study, we conducted an investigative analysis via data-driven evaluation of potential hepatocellular carcinoma (HCC) therapeutics in the context of AI-assisted drug discovery/repurposing. First, we discussed basic concepts, computational approaches, databases, modeling approaches, and featurization techniques in drug discovery/repurposing. In the analysis part, we automatically integrated HCC-related biological entities such as genes/proteins, pathways, phenotypes, drugs/compounds, and other diseases with similar implications, and represented these heterogeneous relationships via a knowledge graph using the CROssBAR system. RESULTS: Following the system-level evaluation and selection of critical genes/proteins and pathways to target, our deep learning-based drug/compound-target protein interaction predictors DEEPScreen and MDeePred have been employed for predicting new bioactive drugs and compounds for these critical targets. Finally, we embedded ligands of selected HCC-associated proteins which had a significant enrichment with the CROssBAR system into a 2-D space to identify and repurpose small molecule inhibitors as potential drug candidates based on their molecular similarities to known HCC drugs. CONCLUSIONS: We expect that these series of data-driven analyses can be used as a roadmap to propose early-stage potential inhibitors (from database-scale sets of compounds) to both HCC and other complex diseases, which may subsequently be analyzed with more targeted in silico and experimental approaches.

Defining a master curve of abdominal aortic aneurysm growth and its potential utility of clinical management.

OBJECTIVE: The maximum diameter measurement of an abdominal aortic aneurysm (AAA), which depends on orthogonal and axial cross-sections or maximally inscribed spheres within the AAA, plays a significant role in the clinical decision making process. This study aims to build a total of 21 morphological parameters from longitudinal CT scans and analyze their correlations. Furthermore, this work explores the existence of a "master curve" of AAA growth, and tests which parameters serve to enhance its predictability for clinical use. METHODS: 106 CT scan images from 25 Korean AAA patients were retrospectively obtained. We subsequently computed morphological parameters, growth rates, and pair-wise correlations, and attempted to enhance the predictability of the growth for high-risk aneurysms using non-linear curve fitting and least-square minimization. RESULTS: An exponential AAA growth model was fitted to the maximum spherical diameter, as the best representative of the growth among all parameters (r-square: 0.94) and correctly predicted to 15 of 16 validation scans based on a 95% confidence interval. AAA volume expansion rates were highly correlated (r=0.75) with thrombus accumulation rates. CONCLUSIONS: The exponential growth model using spherical diameter provides useful information about progression of aneurysm size and enables AAA growth rate extrapolation during a given surveillance period.

Predicting abdominal aortic aneurysm growth using patient-oriented growth models with two-step Bayesian inference.

OBJECTIVE: For small abdominal aortic aneurysms (AAAs), a regular follow-up examination is recommended every 12 months for AAAs of 30-39 mm and every six months for AAAs of 40-55 mm. Follow-up diameters can determine if a patient follows the common growth model of the population. However, the rapid expansion of an AAA, often associated with higher rupture risk, may be overlooked even though it requires surgical intervention. Therefore, the prognosis of abdominal aortic aneurysm growth is clinically important for planning treatment. This study aims to build enhanced Bayesian inference methods to predict maximum aneurysm diameter. METHODS: 106 CT scans from 25 Korean AAA patients were retrospectively obtained. A two-step approach based on Bayesian calibration was used, and an exponential abdominal aortic aneurysm growth model (population-based) was specified according to each individual patient's growth (patient-specific) and morphologic characteristics of the aneurysm sac (enhanced). The distribution estimates were obtained using a Markov Chain Monte Carlo (MCMC) sampler. RESULTS: The follow-up diameters were predicted satisfactorily (i.e. the true follow-up diameter was in the 95% prediction interval) for 79% of the scans using the population-based growth model, and 83% of the scans using the patient-specific growth model. Among the evaluated geometric measurements, centerline tortuosity was a significant (p = 0.0002) predictor of growth for AAAs with accelerated and stable expansion rates. Using the enhanced prediction model, 86% of follow-up scans were predicted satisfactorily. The average prediction errors of population-based, patient-specific, and enhanced models were ±2.67, ±2.61 and ± 2.79 mm, respectively. CONCLUSION: A computational framework using patient-oriented growth models provides useful tools for per-patient basis treatment and enables better prediction of AAA growth.

Real-time Malaria Parasite Screening in Thick Blood Smears for Low-Resource Setting.

Malaria is a serious public health problem in many parts of the world. Early diagnosis and prompt effective treatment are required to avoid anemia, organ failure, and malaria-associated deaths. Microscopic analysis of blood samples is the preferred method for diagnosis. However, manual microscopic examination is very laborious and requires skilled health personnel of which there is a critical shortage in the developing world such as in sub-Saharan Africa. Critical shortages of trained health personnel and the inability to cope with the workload to examine malaria slides are among the main limitations of malaria microscopy especially in low-resource and high disease burden areas. We present a low-cost alternative and complementary solution for rapid malaria screening for low resource settings to potentially reduce the dependence on manual microscopic examination. We develop an image processing pipeline using a modified YOLOv3 detection algorithm to run in real time on low-cost devices. We test the performance of our solution on two datasets. In the dataset collected using a microscope camera, our model achieved 99.07% accuracy and 97.46% accuracy on the dataset collected using a mobile phone camera. While the mean average precision of our model is on par with human experts at an object level, we are several orders of magnitude faster than human experts as we can detect parasites in images as well as videos in real time.

The Use of Informed Priors in Biclustering of Gene Expression with the Hierarchical Dirichlet Process.

We motivate and describe the application of Hierarchical Dirichlet Process (HDP) models to the "soft" biclustering of gene expression data, in which we obtain modules (biclusters) where the affiliation of genes and samples with the modules are weighted, instead of being hard memberships. As a distinct contribution, we propose a method which HDP is informed with prior beliefs, significantly increasing the quality of the biclustering in terms of both the correctness of the number of modules inferred, and the precision of these modules, especially when evidence is sparse. We outline two such informed priors; one based on co-expression relationships inherent in the data, the other based on an externally provided regulatory network. We validate these results and compare the performance of our approach to Weighted Gene Correlation Network Analysis (WGCNA), another model that features weighted modules. We have, to this end, performed experiments on semi-synthetic data. The results show that HDP, with the addition of a well-informed prior, is able to capture the correct number of modules with increased accuracy. Furthermore, the model becomes robust to changes in the strength of the prior. We conclude by discussing these results and the benefits provided by our approach for gene expression analysis and network validation.

mESAdb: microRNA expression and sequence analysis database.

microRNA expression and sequence analysis database (http://konulab.fen.bilkent.edu.tr/mirna/) (mESAdb) is a regularly updated database for the multivariate analysis of sequences and expression of microRNAs from multiple taxa. mESAdb is modular and has a user interface implemented in PHP and JavaScript and coupled with statistical analysis and visualization packages written for the R language. The database primarily comprises mature microRNA sequences and their target data, along with selected human, mouse and zebrafish expression data sets. mESAdb analysis modules allow (i) mining of microRNA expression data sets for subsets of microRNAs selected manually or by motif; (ii) pair-wise multivariate analysis of expression data sets within and between taxa; and (iii) association of microRNA subsets with annotation databases, HUGE Navigator, KEGG and GO. The use of existing and customized R packages facilitates future addition of data sets and analysis tools. Furthermore, the ability to upload and analyze user-specified data sets makes mESAdb an interactive and expandable analysis tool for microRNA sequence and expression data.