Team Fudan 2022 Software Tool - PartHub

This is the mirror repository of Software of Fudan - iGEM 2022.

Improve your part design with 50000+ parts！

Try our public PartHub here!

Description

For more information, please visit our wiki page Software | Fudan - iGEM 2022

Highlight

• A quick, efficient and powerful tool for building and designing new parts
• Provide multiple ways for searching for 50000+ existing parts with Registry of Standard Biological Parts platform
• Visualization of the relationship between parts
• Can be used across various platforms, and intuitive workflow

Installation

If you only want to use the public PartHub, you don't need to perform installation. Please click here and read the how-to section.

Since there is no image of Oracle JDK in the docker hub, it is not convenient to install PartHub using docker. We recommend using Amazon Machine Image (AMI) to install it.

(Recommended) Install with AMI

Quick installation guide

1. Open the Amazon EC2 console at https://console.amazonaws.cn/ec2/.

2. Under Images > AMI Catalog, do the following:

a. Search PartHub in Community AMls

b. Select the PartHub AMI (ami-036510704fd194c8f) and choose Launch Instance with AMI

3. Select Instance type as t3.xlarge or higher and Launch instance.

4. Configure the Inbound rules under Network & Security > Security Groups as follows:

   IP version	Type	Protocol	Port range	Source	Description
   IPv4	Custom TCP	TCP	7687	0.0.0.0/0	neo4j
   IPv4	Custom TCP	TCP	7474	0.0.0.0/0	neo4j
   IPv4	Custom TCP	TCP	5000	0.0.0.0/0	web
   IPv4	SSH	TCP	22	0.0.0.0/0	–

5. Connect to your server using SSH, and run the following in your Terminal:

   screen -S neo4j

   then run the following in the screen:

   sudo -s
   cd /usr/local/neo4j-community-4.4.11/bin/
   ./neo4j console

6. Press Crtl + A and then press D to detach the screen session and run the following in your Terminal:

   screen -S web
   

   then run the following in the screen:

   cd /home/ubuntu/fudan-main/
   source venv/bin/activate
   python /home/ubuntu/fudan-main/DataBase/LoadCSVFile.py
   gunicorn -b 0.0.0.0:5000 app:app

7. Visit http://serverIP:5000/ (change serverIP to the IP of your server) and start using your PartHub!

Manual installation (For local deployment)

The system requirements to deploy PartHub are as follows:

Minimum		Recommended
CPU	2-core, 2.4 GHz	CPU	8-core, 2.9 GHz
Memory	8 GB	Memory	16 GB
Hard Drive	20 GB, SSD	Hard Drive	512 GB, SSD
Network	Broadband Internet connection	Network	Broadband Internet connection

Environmental requirements: Neo4j & Oracle JDK (We recommend Neo4j 4.4.11 & Oracle JDK 11.0.16). For more information, see https://neo4j.com/docs/operations-manual/current/installation/ & https://docs.oracle.com/en/java/javase/11/install/overview-jdk-installation.html. For Neo4j, please set the Username as neo4j & Password as igem2022.

1. Check your JDK version in your Terminal:

java --version

If the JDK version is correct, the return should be:

java 11.0.16 2022-07-19 LTS
Java(TM) SE Runtime Environment 18.9 (build 11.0.16+11-LTS-199)
Java HotSpot(TM) 64-Bit Server VM 18.9 (build 11.0.16+11-LTS-199, mixed mode)

2. Download PartHub:

   [email protected]:2022/software-tools/fudan.git

3. Replace the neo4j.conf in the conf file where you install your Neo4j (e.g. /usr/local/neo4j-community-4.4.11/conf/) with that in this Repository.

4. Change directory to the location of the bin file of Neo4j, and run the following in your Terminal:

   screen -S neo4j

   then run the following in the screen:

   sudo -s
   ./neo4j console

5. Press Crtl + A and then press D to detach the screen session and run the following in your Terminal:

   screen -S web

   then run the following in the screen:

   cd /home/ubuntu/fudan-main/
   python3 -m venv venv
   source venv/bin/activate
   pip install -r requirements.txt
   python /home/ubuntu/fudan-main/DataBase/LoadCSVFile.py
   gunicorn -b 0.0.0.0:5000 app:app
   

   where /home/ubuntu/fudan-main/ should be changed to the directory of your fudan-main file.

6. Visit http://serverIP:5000/ (change serverIP to the IP of your server) and start using your PartHub!

How-to

For common users

Below is an overview of PartHub's workflow.

The process of using PartHub is shown below:

1. Visit PartHub, for example, http://3.238.241.161:5000/. The website is supposed to be like this:

   

2. Select the search type and enter search terms (e.g. search type is Content & search term is carotene). The format of the search terms corresponding to different search types is as follows:

   Search type	Meaning & Format	Example	Example URL
   ID	The id of the part (e.g. BBa_xxxxxxxx). It is fine to enter a search term with or without BBa_.	K3790012	http://3.238.241.161:5000/sp?s=K3790012&searchtype=number
   Name	The name of the part in Registry of Standard Biological Parts.	GFP	http://3.238.241.161:5000/sp?s=GFP&searchtype=name
   Sequence	The sequence of the part (if it exists). The search term can only be the combination of [a, t, g, c, A, T, G, C].	TTAACTTTAAGAAGGAG	http://3.238.241.161:5000/sp?s=TTAACTTTAAGAAGGAG&searchtype=sequence
   Designer	The name of the person who designed the part.	Guanqiao Chi	http://3.238.241.161:5000/sp?s=Guanqiao%20Chi&searchtype=designer
   Team	The name of the team which designed the part.	Fudan	http://3.238.241.161:5000/sp?s=Fudan&searchtype=team
   Content	The content of the part in Registry of Standard Biological Parts.	carotene	http://3.238.241.161:5000/sp?s=carotene&searchtype=contents

   Our search engine is case-insensitive and support partial match retrieval. In addition, PartHub supports boolean search with multiple search terms (The format of the boolean search with multiple search terms should be xxx AND xxx or xxx OR xxx).

   What's more, PartHub supports fuzzy search. For example, if you want to search for beta AND carotene but you accidentally type in bata AND carotene, PartHub will automatically do the fuzzy search for you and the result will be:

   

3. View sorted searching results (you can choose certain sorting order). PartHub provides several ways to sort the results, like Most cited, Best match, etc. In this way, users can organize the result in order to their needs. Sorting by Recommended is a weighted sort that considers various aspects of suitability and can be used in situations where users are unsure which sort is more appropriate for their needs.

4. Click on a specific part you are interested in and visualize the relationship network of this part. The page is as follows:

   

   This relationship network is interactive, you can scroll to zoom the canvas and drag to move the nodes. Click to display the part details and double click to go to the part page. You can drag nodes to change the layout of the canvas. On this page, you can also get the sequence of this part directly.

For developers

The source code is in this Repository.

1. The DataBase file contains code to import data into the database.

2. The WebCrawler file contains code to get data from the Registry of Standard Biological Parts. If you want to update the data for a new year, replace the year in line 679 of software_ver0.4.py with the year you want to get the data for. Then you can run MergeCSVFiles.py to get the new all_collections.csv. Then run Preprocessing.py to get all_collections_filted.csv. After uploading the data via LoadCSVFile.py, you are able to use PartHub with updated data. We recommend you to run it on a PC with more than 8 core CPUs for getting data of just one year, and on a higher performance computer for getting multiple years of data. Remember to switch the appropriate number of threads in software_ver0.4.py for better performance. Our recommended configuration is as follows:

   CPUs	Dual Intel® Xeon E5-2680v2(10core, 20thread, 2.80 GHz to 3.60 GHz)
   Memory	64 GB, DDR3, 1600 MHz
   Hard Drive	500 GB, NVMe, SSD
   GPU	NVIDIA® GeForce GTX1050 Ti

3. The back-end code of PartHub is written in Python (.py files). The role of each code has been commented in the file.

4. The static file contains static files(images and JavaScript files). The templates file contains front-end HTML files.

Contributing

Pull Requests

We actively welcome your pull requests.

1. Fork the repo and create your branch from main.
2. If you've added code that should be tested, add tests.
3. If you've changed APIs, update the documentation.
4. Ensure the test suite passes.
5. Make sure your code lints.

In order to accept your pull request, please send me an email.

Issues

We use GitLab issues to track bugs. Please describe your isuee clearly and give sufficient instructions to reproduce the issue.

Authors and acknowledgment

Authors

• Zhiyue Chen (@mistyfield)
• Yunjia Liu (@VeronicaLiu)

Acknowledgment

• Sunzhe Kang (@KangSunzhe)
• Shitao Gong (@Tom_GhoST_Smith), and 2021's Part Camera