A5028196556- RNA modifications and cancer
- Cancer-related molecular mechanisms research
- Genomics and Phylogenetic Studies
- Bioinformatics and Genomic Networks
- MicroRNA in disease regulation
- RNA and protein synthesis mechanisms
- RNA Research and Splicing
- Computational Drug Discovery Methods
- Gene expression and cancer classification
- Music and Audio Processing
- Topic Modeling
- Advanced Text Analysis Techniques
- Speech and Audio Processing
- Single-cell and spatial transcriptomics
- Neurobiology of Language and Bilingualism
- Gene Regulatory Network Analysis
- vaccines and immunoinformatics approaches
- Biomedical Text Mining and Ontologies
- T-cell and Retrovirus Studies
- Natural Language Processing Techniques
- Music Technology and Sound Studies
- Dementia and Cognitive Impairment Research
- Cognitive Functions and Memory
- Frailty in Older Adults
- Remote-Sensing Image Classification
Chulalongkorn University
2010-2025
Genome Institute of Singapore
2020-2022
Agency for Science, Technology and Research
2020-2022
European Bioinformatics Institute
2016-2017
Open Targets
2016
Wellcome Trust
2016
University of Cambridge
2014-2015
We have designed and developed a data integration visualization platform that provides evidence about the association of known potential drug targets with diseases. The is to support identification prioritization biological for follow-up. Each target linked disease using integrated genome-wide from broad range sources. either target-centric workflow identify diseases may be associated specific target, or disease-centric disease. Users can easily transition between these target- workflows....
Abstract RNA modifications such as m6A methylation form an additional layer of complexity in the transcriptome. Nanopore direct sequencing can capture this information raw current signal for each molecule, enabling detection using supervised machine learning. However, experimental approaches provide only site-level training data, whereas modification status single molecule is missing. Here we present m6Anet, a neural-network-based method that leverages multiple instance learning framework to...
Single-cell RNA-sequencing (scRNA-seq) allows studying heterogeneity in gene expression large cell populations. Such can arise due to technical or biological factors, making decomposing sources of variation difficult. We here describe f-scLVM (factorial single-cell latent variable model), a method based on factor analysis that uses pathway annotations guide the inference interpretable factors underpinning heterogeneity. Our model jointly estimates relevance individual refines set...
Abstract The human genome contains more than 200,000 gene isoforms. However, different isoforms can be highly similar, and with an average length of 1.5kb remain difficult to study short read sequencing. To systematically evaluate the ability transcriptome at a resolution individual we profiled 5 cell lines cDNA sequencing Nanopore long direct RNA, amplification-free cDNA, PCR-cDNA protocols showed high level consistency, RNA being most similar. While reads generated comparable expression...
Abstract The human genome contains instructions to transcribe more than 200,000 RNAs. However, many RNA transcripts are generated from the same gene, resulting in alternative isoforms that highly similar and remain difficult quantify. To evaluate ability study transcript expression, we profiled seven cell lines with five different RNA-sequencing protocols, including short-read cDNA, Nanopore long-read direct RNA, amplification-free cDNA PCR-amplified sequencing, PacBio IsoSeq, multiple...
The Montreal cognitive assessment (MoCA), a widely accepted screening tool for identifying patients with mild impairment (MCI), includes language fluency test of verbal functioning; its scores are based on the number unique correct words produced by taker. However, it is possible that may be counted differently various languages. This study focuses Thai as differs from English in terms word combinations. We applied automatic speech recognition (ASR) techniques to develop an assisted scoring...
Abstract RNA modifications such as m6A methylation form an additional layer of complexity in the transcriptome. Nanopore direct sequencing captures this information raw current signal for each molecule, enabling detection using supervised machine learning. However, experimental approaches provide only site-level training data, whereas modification status single molecule is missing. Here we present m6Anet, a neural network-based method that leverages Multiple Instance Learning framework to...
Several COVID-19 vaccination rollout strategies are implemented. Real-world data from the large-scale, government-mandated Central Vaccination Center (CVC), Thailand, could be used for comparing breakthrough infection, across all available profiles.This prospective cohort study combined vaccine profiles CVC registry with three nationally validated outcome datasets to assess hospitalization, and death among Thais individuals who received at least one dose of vaccine. The outcomes were...
The growing availability of multiomic data provides a highly comprehensive view cellular processes at the levels mRNA, proteins, metabolites, and reaction fluxes. However, due to probabilistic interactions between components depending on environment time course, casual, sometimes rare may cause important effects in physiology. To date, pathway level cannot be measured directly, methodologies predict cross-correlations from fluxes are still missing. Here, we develop approach flux-balance...
Drug treatments often perturb the activities of certain pathways, sets functionally related genes. Examining pathways/gene that are responsive to drug instead a simple list regulated genes can advance our understanding about such cellular processes after perturbations. In general, pathways do not work in isolation and their connections cause secondary effects. To address this, we present new method better identify pathway responsiveness simultaneously determine between-pathway interactions....
Abstract Differences in RNA expression can provide insights into the molecular identity of a cell, pathways involved human diseases, and variation levels across patients associated with clinical phenotypes. modifications such as m6A have been found to contribute functions RNAs. However, quantification differences has challenging. Here we develop computational method (xPore) identify differential from direct sequencing data. We evaluate our on transcriptome-wide profiling data, demonstrating...
In this paper, we analyze the housing price data obtained from a leading Thai real estate website and Open Street Maps (OSM) to identify features that affect in Thailand 2015 2019. Moreover, propose model based on stacking ensemble learning framework, where predictions are generated by three base models consisting of convolutional neural network (CNN), an (such as random forests (RF), extreme gradient boosting (XGBoost) adaptive (AdaBoost)) simple linear regression technique. The CNN is used...
By representing large corpora with concise and meaningful elements, topic-based generative models aim to reduce the dimension understand content of documents. Those techniques originally analyze on words in documents, but their extensions currently accommodate meta-data such as authorship information, which has been proved useful for textual modeling. The importance learning is extract author interests assign authors anonymous texts. Author-Topic (AT) model, an unsupervised technique,...
Single-cell RNA-sequencing (scRNA-seq) allows heterogeneity in gene expression levels to be studied large populations of cells. Such can arise from both technical and biological factors, thus making decomposing sources variation extremely difficult. We here describe a computationally efficient model that uses prior pathway annotation guide inference the drivers underpinning heterogeneity. Moreover, we jointly update improve set infer factors explaining variability fall outside existing...
This case study uses the Karate Framework to reduce time and resources required for testing an application programming interface (API) Online Assessment Web Application, which is part of a continuous development process involving numerous complex features test cases challenging perform tests manually. The aims verify correctness API's operation facilitate regression during promptly detect errors. involved 162 using Framework. Additionally, software quality assurance reports were created...
Mild cognitive impairment (MCI) is an early stage of decline or memory loss, commonly found among the elderly. A phonemic verbal fluency (PVF) task a standard test that participants are asked to produce words starting with given letters, such as “F” in English and “ก” /k/ Thai. With state-of-the-art machine learning techniques, features extracted from PVF data have been widely used detect MCI. The features, including acoustic semantic word grouping, studied many languages but not However,...
Text summarization is a process of condensing lengthy texts while preserving their essential information. Previous studies have predominantly focused on high-resource languages, low-resource languages like Thai received less attention. Furthermore, earlier extractive models for primarily relied the article's body, without considering headline. This omission can result in exclusion key sentences from summary. To address these limitations, we propose CHIMA, an model that incorporates...
Music genre classification is a widely researched topic in music information retrieval (MIR). Being able to automatically tag genres will benefit streaming service providers such as JOOX, Apple Music, and Spotify for their content-based recommendation. However, most studies on have been done western songs which differ from Thai songs. Lukthung, distinctive long-established type of music, one the popular Thailand has specific group listeners. In this paper, we develop neural networks classify...
When the COVID-19 pandemic began, world had to step into a new normal world, forcing real-life activities transform online activities. Formal teaching, learning, and examination were also affected. The learning assessment system is highly complex because it must be able handle mass of students taking exams simultaneously, especially during midterm final exam week. Therefore, have an acceptable response time high reliability. Our been developed utilizing MySQL as database, but was...