Gene2Vec: Distributed Representation of Genes Based on Co-Expression

Du, Jingcheng; Jia, Peilin; Dai, Yulin; Tao, Cui; Zhao, Zhongming; Zhi, Degui

doi:10.1101/286096

Cited by 29 publications

(45 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Both developmental brain gene expression profiles and RNA transcript sequence compositions were used as features for model construction. To reduce the high dimensionality of the input features, which might cause model overfitting, an autoencoder network, gene embedding [ 26 ], and RF-based feature selection were tested. Lastly, we utilized the models to predict and prioritize ASD-associated candidate lncRNAs, which might provide a good list of targets for further investigation.…”

Section: Introductionmentioning

confidence: 99%

Prediction and prioritization of autism-associated long non-coding RNAs using gene expression and sequence features

Wang

2020

BMC Bioinformatics

View full text Add to dashboard Cite

Background Autism spectrum disorders (ASD) refer to a range of neurodevelopmental conditions, which are genetically complex and heterogeneous with most of the genetic risk factors also found in the unaffected general population. Although all the currently known ASD risk genes code for proteins, long non-coding RNAs (lncRNAs) as essential regulators of gene expression have been implicated in ASD. Some lncRNAs show altered expression levels in autistic brains, but their roles in ASD pathogenesis are still unclear. Results In this study, we have developed a new machine learning approach to predict candidate lncRNAs associated with ASD. Particularly, the knowledge learnt from protein-coding ASD risk genes was transferred to the prediction and prioritization of ASD-associated lncRNAs. Both developmental brain gene expression data and transcript sequence were found to contain relevant information for ASD risk gene prediction. During the pre-training phase of model construction, an autoencoder network was implemented for a representation learning of the gene expression data, and a random-forest-based feature selection was applied to the transcript-sequence-derived k-mers. Our models, including logistic regression, support vector machine and random forest, showed robust performance based on tenfold cross-validations as well as candidate prioritization with hypothetical loci. We then utilized the models to predict and prioritize a list of candidate lncRNAs, including some reported to be cis-regulators of known ASD risk genes, for further investigation. Conclusions Our results suggest that ASD risk genes can be accurately predicted using developmental brain gene expression data and transcript sequence features, and the models may provide useful information for functional characterization of the candidate lncRNAs associated with ASD.

show abstract

Section: Introductionmentioning

confidence: 99%

Prediction and prioritization of autism-associated long non-coding RNAs using gene expression and sequence features

Wang

2020

BMC Bioinformatics

View full text Add to dashboard Cite

show abstract

“…Specifically, bioVec [1] and seq2vec [5] have applied the word2vec technique to biological sequences. Similarly, Gene2vec [3], [21], and Dna2vec [13] applied the same technique to gene embedding, protein embedding, and DNA sequence embedding respectively. All these works were based on word2vec, and their k-mer sizes span from 3-8 (dna2vec embed k-mers of length 3 to 8, others work on k-mer size of 3 ).…”

Section: Related Workmentioning

confidence: 99%

A Vector Representation of DNA Sequences Using Locality Sensitive Hashing

Shi

Chen

2019

Preprint

View full text Add to dashboard Cite

Drawing from the analogy between natural language and "genomic sequence language", we explored the applicability of word embeddings in natural language processing (NLP) to represent DNA reads in Metagenomics studies. Here, k-mer is the equivalent concept of word in NLP and it has been widely used in analyzing sequence data. However, directly replacing word embedding with k-mer embedding is problematic due to two reasons: First, the number of k-mers is many times of the number of words in NLP, making the model too big to be useful. Second, sequencing errors create lots of rare k-mers (noise), making the model hard to be trained. In this work, we leverage Locality Sensitive Hashing (LSH) to overcoming these challenges. We then adopted the skip-gram with negative sampling model to learn k-mer embeddings. Experiments on metagenomic datasets with labels demonstrated that LSH can not only accelerate training time and reduce the memory requirements to store the model, but also achieve higher accuracy than alternative methods. Finally, we demonstrate the trained low-dimensional k-mer embeddings can be potentially used for accurate metagenomic read clustering and predict their taxonomy, and this method is robust on reads with high sequencing error rates (12-22%). KEYWORDSsequence clustering, neural networks, word embedding, LSH

show abstract

“…This technique has been applied to a variety of subjects in biomedical sciences with interesting results. [16][17][18][19][20][21][22] A hallmark feature of word2vec is that the resulting embeddings encode semantic relationships. A classic example would be that, in embeddings trained on an English language corpus, the vectors going from countries to their capital are similar.…”

Section: Representation Of Medication Order Sequences As Word2vec Embmentioning

confidence: 99%

An application of machine learning to assist medication order review by pharmacists in a health care center

Thibault

Lebel

2019

Preprint

View full text Add to dashboard Cite

The objective of this study was to determine if it is feasible to use machine learning to evaluate how a medication order is contextually appropriate for a patient, in order to assist order review by pharmacists. A neural network was constructed using as input the sequence of word2vec embeddings of the 30 previous orders, as well as the currently active medications, pharmacological classes and ordering department, to predict the next order. The model was trained with data from 2013 to 2017, optimized using 5-fold cross-validation, and tested on orders from 2018. A survey was developed to obtain pharmacist ratings on a sample of 20 orders, which were compared with predictions. The training set included 1 022 272 orders. The test set included 95 310 orders. Baseline training set top 1, top 10 and top 30 accuracy using a dummy classifier were respectively 4.5%, 23.6% and 44.1%. Final test set accuracies were, respectively, 44.4%, 69.9% and 80.4%. Populations in which the model performed the best were obstetrics and gynecology patients and newborn babies (either in or out of neonatal intensive care). Pharmacists agreed poorly on their ratings of sampled orders with a Fleiss kappa of 0.283. The breakdown of metrics by population showed better performance in patients following less variable order patterns, indicating potential usefulness in triaging routine orders to less extensive pharmacist review. We conclude that machine learning has potential for helping pharmacists review medication orders. Future studies should aim at evaluating the clinical benefits of using such a model in practice.

show abstract

Gene2Vec: Distributed Representation of Genes Based on Co-Expression

Abstract: 1Background: Existing functional description of genes are categorical, discrete, and mostly

Cited by 29 publications

References 8 publications

Prediction and prioritization of autism-associated long non-coding RNAs using gene expression and sequence features

Prediction and prioritization of autism-associated long non-coding RNAs using gene expression and sequence features

A Vector Representation of DNA Sequences Using Locality Sensitive Hashing

An application of machine learning to assist medication order review by pharmacists in a health care center

Contact Info

Product

Resources

About