Name: a virtual machine platform for non-computer professionals for using deep learning to classify biological sequences of metagenomic data
Uploaded: 2021-09-25T14:00:00
Description: Watch this Scientific Journal Video about A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data at JoVE.com

This investigation was financially supported by the National Natural Science Foundation of China (81925026, 82002201, 81800746, 82102508).

1. The installation of the virtual machine
<ol>
	<li>Download the virtual machine file from (https://github.com/zhenchengfang/DL-VM).</li>
	<li>Download the VirtualBox software from&#160;https://www.virtualbox.org.</li>
	<li>Decompress the &#34;.7z&#34; file using related software, such as &#34;7-Zip&#34;, &#34;WinRAR&#34; or &#34;WinZip&#34;.</li>
	<li>Install the VirtualBox software by clicking the Next button in each step.</li>
	<li>Open the VirtualBox software and click the New but...

<ol>
	<li>Liang, Q., Bible, P. W., Liu, Y., Zou, B., Wei, L. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=DeepMicrobes:+taxonomic+classification+for+metagenomics+with+deep+learning.">DeepMicrobes: taxonomic classification for metagenomics with deep learning.</a> NAR Genomics and Bioinformatics. 2 (1), (2020).</li><li>Ren, J., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=VirFinder:+a+novel+k+-mer+based+tool+for+identifying+viral+sequences+from+assembled+metagenomic+data.">VirFinder: a novel k -mer based tool for identifying viral sequences from assembled metagenomic data.</a> Microbiome. 5 (1), 69 (2017).</li><li>Fang, Z., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=PPR-Meta:+a+tool+for+identifying+phages+and+plasmids+from+metagenomic+fragments+using+deep+learning.">PPR-Meta: a tool for identifying phages and plasmids from metagenomic fragments using deep learning.</a> GigaScience. 8 (6), (2019).</li><li>Ren, J., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Identifying+viruses+from+metagenomic+data+using+deep+learning.">Identifying viruses from metagenomic data using deep learning.</a> Quantitative Biology. 8 (1), 64-77 (2020).</li><li>Zhou, F., Xu, Y. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=cBar:+a+computer+program+to+distinguish+plasmid-derived+from+chromosome-derived+sequence+fragments+in+metagenomics+data.">cBar: a computer program to distinguish plasmid-derived from chromosome-derived sequence fragments in metagenomics data.</a> Bioinformatics. 26 (16), 2051-2052 (2010).</li><li>Krawczyk, P. S., Lipinski, L., Dziembowski, A. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=PlasFlow:+predicting+plasmid+sequences+in+metagenomic+data+using+genome+signatures.">PlasFlow: predicting plasmid sequences in metagenomic data using genome signatures.</a> Nucleic Acids Research. 46 (6), (2018).</li><li>Pellow, D., Mizrahi, I., Shamir, R. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=PlasClass+improves+plasmid+sequence+classification.">PlasClass improves plasmid sequence classification.</a> PLOS Computational Biology. 16 (4), (2020).</li><li>Arango-Argoty, G., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=DeepARG:+a+deep+learning+approach+for+predicting+antibiotic+resistance+genes+from+metagenomic+data.">DeepARG: a deep learning approach for predicting antibiotic resistance genes from metagenomic data.</a> Microbiome. 6 (1), 1-15 (2018).</li><li>Zheng, D., Pang, G., Liu, B., Chen, L., Yang, J. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Learning+transferable+deep+convolutional+neural+networks+for+the+classification+of+bacterial+virulence+factors.">Learning transferable deep convolutional neural networks for the classification of bacterial virulence factors.</a> Bioinformatics. 36 (12), 3693-3702 (2020).</li><li>LeCun, Y., Bengio, Y., Hinton, G. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Deep+learning.">Deep learning.</a> Nature. 521 (7553), 436-444 (2015).</li><li>Fang, Z., Zhou, H. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=VirionFinder:+Identification+of+Complete+and+Partial+Prokaryote+Virus+Virion+Protein+From+Virome+Data+Using+the+Sequence+and+Biochemical+Properties+of+Amino+Acids.">VirionFinder: Identification of Complete and Partial Prokaryote Virus Virion Protein From Virome Data Using the Sequence and Biochemical Properties of Amino Acids.</a> Frontiers in Microbiology. 12, 615711 (2021).</li><li>Fang, Z., Zhou, H. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Identification+of+the+conjugative+and+mobilizable+plasmid+fragments+in+the+plasmidome+using+sequence+signatures.">Identification of the conjugative and mobilizable plasmid fragments in the plasmidome using sequence signatures.</a> Microbial Genomics. 6 (11), (2020).</li><li>Richter, D. C., Ott, F., Auch, A. F., Schmid, R., Huson, D. H. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=MetaSim-a+sequencing+simulator+for+genomics+and+metagenomics.">MetaSim-a sequencing simulator for genomics and metagenomics.</a> PLoS One. 3 (10), 3373 (2008).</li><li>Zhang, M., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Prediction+of+virus-host+infectious+association+by+supervised+learning+methods.">Prediction of virus-host infectious association by supervised learning methods.</a> BMC Bioinformatics. 18 (3), 143-154 (2017).</li></ol>

This tutorial provides an overview for biologists and algorithm design beginners on how to construct an easy-to-use deep learning framework for biological sequence classification in metagenomic data. This tutorial aims to provide intuitive understanding of deep learning and address the challenge that beginners often have difficulty installing the deep learning package and writing the code for the algorithm. For some simple classification tasks, users can use the framework to perform the classification tasks.
<p class...

The metagenomic sequencing technique bypasses the strain isolation process and directly sequences the total DNA in an environmental sample. Thus, metagenomic data contain DNA from different organisms, and most biological sequences are from novel organisms that are not present in the current database. According to different research purposes, biologists need to classify these sequences from different perspectives, such as taxonomic classification1, virus-bacteria classification2,3,4, chromosome-plasmid classification3,5,6,7, and gene function annotation (such as antibiotic resistance gene classification8 and virulence factor classification9). Because metagenomic data contain a large number of novel species and genes, ab initio algorithms, which do not rely on known databases for sequence classification (including DNA classification and protein classification), are an important approach in metagenomic data analysis. However, the design of such algorithms requires professional mathematics knowledge and programming skills; therefore, many biologists and algorithm design beginners have difficulty constructing a classification algorithm to suit their own needs.
With the development of artificial intelligence, deep learning algorithms have been widely used in the field of bioinformatics to complete tasks such as sequence classification in metagenomic analysis. To help beginners understand deep learning algorithms, we describe the algorithm in an easy-to-understand fashion below.
An overview of a deep learning technique is shown in Figure 1. The core technology of a deep learning algorithm is an artificial neural network, which is inspired by the structure of the human brain. From a mathematical point of view, an artificial neural network may be regarded as a complex function. Each object (such as a DNA sequence, a photo or a video) is first digitized. The digitized object is then imported to the function. The task of the artificial neural network is to give a correct response according to the input data. For example, if an artificial neural network is constructed to perform a 2-class classification task, the network should output a probability score that is between 0-1 for each object. The neural network should give the positive object a higher score (such as a score higher than 0.5) while giving the negative object a lower score. To obtain this goal, an artificial neural network is constructed with the training and testing processes. During these processes, data from the known database are downloaded and then divided into a training set and test set. Each object is digitized in a proper way and given a label (&#34;1&#34; for positive objects and &#34;0&#34; for negative objects). In the training process, the digitized data in the training set are inputted into the neural network. The artificial neural network constructs a loss function that represents the dissimilarity between the output score of the input object and the corresponding label of the object. For example, if the label of the input object is &#34;1&#34; while the output score is &#34;0.1&#34;, the loss function will be high; and if the label of the input object is &#34;0&#34; while the output score is &#34;0.1&#34;, the loss function will be low. The artificial neural network employs a specific iterative algorithm that adjusts the parameters of the neural network to minimize the loss function. The training process finishes when the loss function cannot be obviously further decreased. Finally, the data in the test set are used to test the fixed neural network, and the ability of the neural network to calculate the correct labels for the novel objects is evaluated. More principles of deep learning algorithms can be found in the review in LeCun et al.10.
Although the mathematical principles of deep learning algorithms may be complex, many highly packaged deep learning packages have recently been developed, and programmers can directly construct a simple artificial neural network with a few lines of code.
To assist biologists and algorithm design beginners in getting started in using deep learning more quickly, this tutorial provides a guideline for constructing an easy-to-use deep learning framework for sequence classification. This framework uses the &#34;one-hot&#34; encoding form as the mathematical model to digitize the biological sequences and uses a convolution neural network to perform the classification task (see the Supplementary Material). The only thing that the users need to do before using this guideline is to prepare four sequence files in &#34;fasta&#34; format. The first file contains all sequences of the positive class for the training process (referred to &#34;p_train.fasta&#34;); the second file contains all sequences of the negative class for the training process (referred to &#34;n_train.fasta&#34;); the third file contains all sequences of the positive class for the testing process (referred to &#34;p_test.fasta&#34;); and the last file contains all sequences of the negative class for the testing process (referred to &#34;n_test.fasta&#34;). The overview of the flowchart of this tutorial is provided in Figure 2, and more details will be mentioned below.

<table border="1" fo:keep-together.within-page="1" fo:keep-with-next.within-page="always">
	<tbody>
		<tr>
			<td>PC or server</td>
			<td>NA</td>
			<td>NA</td>
			<td>Suggested memory: &#62;6GB</td>
		</tr>
		<tr>
			<td>VirtualBox software</td>
			<td>NA</td>
			<td>NA</td>
			<td>Link: https://www.virtualbox.org</td>
		</tr>
	</tbody>
</table>

a virtual machine platform for non-computer professionals for using deep learning to classify biological sequences of metagenomic data

A variety of biological sequence classification tasks, such as species classification, gene function classification and viral host classification, are expected processes in many metagenomic data analyses. Since metagenomic data contain a large number of novel species and genes, high-performing classification algorithms are needed in many studies. Biologists often encounter challenges in finding suitable sequence classification and annotation tools for a specific task and are often not able to construct a corresponding algorithm on their own because of a lack of the necessary mathematical and computational knowledge. Deep learning techniques have recently become a popular topic and show strong advantages in many classification tasks. To date, many highly packaged deep learning packages, which make it possible for biologists to construct deep learning frameworks according to their own needs without in-depth knowledge of the algorithm details, have been developed. In this tutorial, we provide a guideline for constructing an easy-to-use deep learning framework for sequence classification without the need for sufficient mathematical knowledge or programming skills. All the code is optimized in a virtual machine so that users can directly run the code using their own data.

In our previous work, we developed a series of sequence classification tools for metagenomic data using an approach similar to this tutorial3,11,12. As an example, we deposited the sequence files of the subset of training set and test set from our previous work3,11 in the virtual machine.
Fang &#38; Zhou11 aimed to iden...

Watch this Scientific Journal Video about A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data at JoVE.com

A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data

This tutorial describes a simple method to construct a deep learning algorithm for performing 2-class sequence classification of metagenomic data.

A variety of biological sequence classification tasks, such as species classification, gene function classification and viral host classification, are ...

A variety of biological sequence classification task, such as species classification, gene function classification, and wire host classification are expected processes in many metagenomic data analyzes. Since metagenomic data contain a large number of Novo species and genes, high performing classification organisms are needed in many studies. Biologists often encounter challenge in finding suitable sequence classification and notation tools for a specific task and are often not able to construct a corresponding organism on their own because of a lack of the necessary mathematical and computational knowledge.Deep learning techniques have recently become a popular topic and show strong advantage in many classification tasks. To date, many highly packaged deep learning package, which make it possible for biologists to construct deep learning frameworks, according to their own needs without in depth knowledge of the organism details have been developed. In this tutorial, we provide a guideline for constructing an easy to use deep learning framework for sequence classification without the need for sufficient mathematical knowledge or programming skills.The following video shows how to use the virtual machine to perform biological sequence classification. Users need to download the virtual machine file from the tutorial homepage, and then download the VirtualBox software. The virtual machine is compressed as a seventy file.The seventy file can easily be decompressed using a current compressing software, such as WinRar, Winzip, and 7-Zip. We decompressed the virtual machine using 7-Zip. The decompression may take some time.Please wait for awhile. After decompression users need to install the VirtualBox software. Create a folder to install the VirtualBox.Create a VirtualBox installation package. Select the folder created by yourself. Then install the VirutalBox software by clicking the next button in each step.The installation may take some time, please wait for awhile. Open the VirtualBox software. Create a new button to create a virtual machine.Enter the virtual machine name specified by yourself in the name frame. Select Linux as the operating system in the type frame. Select Ubuntu in the version frame and click the next button.If possible, allocate a larger amount of memory to the virtual machine. True the use an existing hard disk file selection. Select the virtual machine file downloaded from the tutorial homepage.And then click the create button. Click a start button to open the virtual machine. Starting up the virtual machine may take a while.Please wait for a moment before the next step. Then users need to create shared folder in both physical hosts and virtual machine to exchange files. In your physical host, create a shared folder named shared host and on the desktop of the virtual machine, create a shared folder named shared VM.In the manual bar of the virtual machine, click devices, shared folders, shared folder settings successively.Click the button in the upper right corner. Select the shared folder in the physical host created by yourself. Select the auto mount option.Click the OK button. Then restart the virtual machine. Restarting the virtual machine may take awhile.Please wait for a moment before the next step. Click the right click on the desktop of the virtual machine and open the terminal. Type the following command to the terminal.Sudo, space key, mount, space key, bar T, space key, vboxsf, space key, shared host, space key, dot slash, desktop, slash, shared VM.When prompted for a password, enter one and tap the enter key. Copy all four sequence files in faster format for the training and testing process to the shared host folder of the physical host. In this way, all the files will also occur in the shared VM folder of the virtual machine.Then copy the files in the shared VM folder to the deep learning folder of the virtual machine. Click the right click and open the terminal and type the following command to perform the one hot encoding. Dot slash, one hot encoding, specify the files for training and testing.And specify the sequence type. Then type the following command to start the trending process. Python space key, train dot P Y.Then the trending process will begin.This process may take a few hours or a few days, depending on your data set size. When the process is finished, the predict result of the test data is present in the predict dot CSV file. In our previous work, we developed a series of sequence classification tools for a metagenomic data, using an approach similar to this tutorial.For example, we developed a tool aimed to identify the complete and partial prokaryote virus virion proteins from run data. And a tool aimed to identify phage DNA fragments from bacterial chromosome DNA fragments in metogenomic data. The performance of the tools using the script of this tutorial is shown in the figure a and b.In conclusion, this tutorial provides an overview for biologist and organisms design beginners on how to construct an easy to use deep learning framework for biological sequence classification in metogenomic data. This tutorial aims to provide intuitive understanding of deep learning and address the challenge that beginners often have difficulty in starting the deep learning package and writing the code for the organism. For some simple classification tasks, users can use our framework to perform the classification task.

Research

JoVE Journal

Genetics

A Robotic Platform for High-throughput Protoplast Isolation and Transformation

Over the last decade there has been a resurgence in the use of plant protoplasts that range from model species to crop species, for analysis of signal ...

Perturbations of Circulating miRNAs in Irritable Bowel Syndrome Detected Using a Multiplexed High-throughput Gene Expression Platform

The gene expression platform assay allows for robust and highly reproducible quantification of the expression of up to 800 transcripts (mRNA or miRNAs) in ...

Metagenomic Analysis of Silage

Metagenomics is defined as the direct analysis of deoxyribonucleic acid (DNA) purified from environmental samples and enables taxonomic identification of ...

Using a GFP-tagged TMEM184A Construct for Confirmation of Heparin Receptor Identity

When novel proteins are identified through affinity-based isolation and bioinformatics analysis, they are often largely uncharacterized. Antibodies ...

Protocols for C-Brick DNA Standard Assembly Using Cpf1

CRISPR-associated protein Cpf1 cleaves double-stranded DNA under the guidance of CRISPR RNA (crRNA), generating sticky ends. Because of this ...

Robust DNA Isolation and High-throughput Sequencing Library Construction for Herbarium Specimens

Herbaria are an invaluable source of plant material that can be used in a variety of biological studies. The use of herbarium specimens is associated with ...

CRISPR/Cas9-mediated Targeted Integration In Vivo Using a Homology-mediated End Joining-based Strategy

CRISPR/Cas9-mediated Targeted Integration In Vivo Using a Homology-mediated End Joining-based Strategy

As a promising genome editing platform, the CRISPR/Cas9 system has great potential for efficient genetic manipulation, especially for targeted integration ...

Informatic Analysis of Sequence Data from Batch Yeast 2-Hybrid Screens

We have adapted the yeast 2-hybrid assay to simultaneously uncover dozens of transient and static protein interactions within a single screen utilizing ...

High-throughput Identification of Gene Regulatory Sequences Using Next-generation Sequencing of Circular Chromosome Conformation Capture (4C-seq)

High-throughput Identification of Gene Regulatory Sequences Using Next-generation Sequencing of Circular Chromosome Conformation Capture 4C-seq

The identification of regulatory elements for a given target gene poses a significant technical challenge owing to the variability in the positioning and ...

A Deep-sequencing-assisted, Spontaneous Suppressor Screen in the Fission Yeast Schizosaccharomyces pombe

A Deep-sequencing-assisted, Spontaneous Suppressor Screen in the Fission Yeast Schizosaccharomyces pombe

A genetic screen for mutant alleles that suppress phenotypic defects caused by a mutation is a powerful approach to identify genes that belong to closely ...