• Rlscoper

    Just Like That

    We provide a very friendly software to mine informations of RNA-RNA interaction and Protein-Protein interaction .

    Download

Hello people! Welcome to our Rlscoper!

From this website,you can do as follows:

Introduction

The explosive growth of the crosstalk between protein-coding RNAs and non-coding RNAs has been revealed as one of reasons that unexpected complexity of eukaryote transcriptome. In recent years, the expanding transcriptomic researches have brought the RNA-RNA interaction back into people's sight again. The RNA-RNA interaction has been as important as the Protein-Protein interaction. However, mining RNA-RNA interactions by manually reading will waste lots of time and manpower and material resources. Thus, we have developed PRI Finder, a friendly software based on the core algorithm of N-gram statistical language model. Users can use this software to get informations of RNA-RNA interaction and Protein-Protein interaction from abstracts or full texts of literatures. In addition, PRI Finder also provides the literature search online and downloading automatically.

Download the PRIF software

Download the PRIF software

PRI Finder can mine informations of RNA-RNA interaction and Protein-Protein interaction .

Explore   the   RAID Database

Explore   the   RAID Database

RAID is the only one database describing the relationship of RNA-RNA interaction by manual collection .

View     the     Documentation

View      the      Documentation

In this part, there are User Guide, Tutorial and FAQ. These can help you know how to use the PRI Finder software better .

 
 

Hello people! Welcome to our PRI Finder!

Main Page

The navigation on the left displays documentation on PRI Finder software.
If you are new to PRI Finder, or you have a question, see the FAQ or the tutorial. The tutorial provides a brief overview of the software, and describes how to prepare input data files, load data files, run the PRI Finder software, and export the result.
If you are interested in the source code of the PRI Finder software, see Software License for getting download.
If the operating system of your computer is not windows, see For Broad Users, which provides other version of the software adapting other operating systems.
If you have comments or questions not answered by the content of left navigation, please contact us.

PRI Finder Algorithm

PRI Finder is a tool to mine the informations of RNA-RNA interaction and Protein-Protein interaction from the abstract or the full text of some literature.
The algorithm is described in the following paper and supplementary information:
Paper-1 Download
Paper-2 Download

Tutorial

  • Overview
  • The PRI Finder Desktop Application Tutorial provides a brief overview of the main features of the PRI Finder application, and describes how to prepare input data files, load data files, run the PRI Finder software, and export the result.

  • Introduction
  • The explosive growth of the crosstalk between protein-coding RNAs and non-coding RNAs has been revealed as one of reasons that unexpected complexity of eukaryote transcriptome. In recent years, the expanding transcriptomic researches have brought the RNA-RNA interaction back into people's sight again. The RNA-RNA interaction has been as important as the Protein-Protein interaction. However, mining RNA-RNA interactions by manually reading will waste lots of time and manpower and material resources. Thus, we have developed PRI Finder, a friendly software based on the core algorithm of N-gram statistical language model. Users can use this software to get informations of RNA-RNA interaction and Protein-Protein interaction from abstracts or full texts of literatures. In addition, PRI Finder also provides the literature search online and downloading automatically.

  • Way to Run PRI Finder
  • The PRI Finder desktop application provides an easy-to-use interface. When you launch the application from the download page or the download button in the Home page of the PRI Finder web site, you should be on windows system to download, install, and start the application.

  • Launching PRI Finder
  • To launch PRI Finder:

  • Go to the Downloads page or get the download button in the Home page.

  • Click the "Click here" or "Download" button to start the PRI Finder Desktop Application.

  • Install the setup package as the guide of the application.

  • Starting PRI Finder
  • When PRI Finder starts, the main window appears. There are four main components of the user interface.

  • The part one provides five input modes.

  • Single Pubmed ID allows the user to input only one pubmed ID;

    Pubmed ID List allows the user to input one txt format file, whose content is pubmed ID list;

    Single Text allows the user to input one txt format file, whose content is the abstract or the full text of one literature;

    Text Directory allows the uset to input one folder, in which there are series of txt format files;

    The last one is Download Pubmed ID List from NCBI with key words and the time range searching, which allows the user to download abstracts from NCBI directly with the software of PRI Finder.

  • The part two is mainly used to set up series of parameters. The threshold is expressed by the number 1 to 20, which means that some degree of false positive. Via the ten-fold cross validation, the results showed that the threshold 5 is the most suitable for using.

  • This part also provides two manners to display results, the way to display all the sentences and the way to display sentences whose scores are in the range of threshold.

    In this part, there are two modes to handle the input datas, processing while downloading and processing after downloading.

    Certainly, there also provides the choosing of cache directory when you choose the mode of handling after downloading.

  • The part three provides the window of displaying results. You can get informations about each sentence, like pubmed ID, score and the original sentence. It is worth mentioning that it can show the full sentence when you click the position of some sentence.

  • As results coming, sentences have three colors in total. It will be red when the score is in the range of the threshold. While, out of the range, and greater than zero, the colour is yellow. At last, when the score is equal to zero, it is green. If you want to study these sentences conveniently, the software also provides the function of exporting results in excel format, with the score and the flag (No colour=0, Yellow=1, Red=2) of each sentence.

  • The part four is Massage, which displays the processing when running PRI Finder software.

  • The user interface provides another function that is the top part and the tail part of the interface can be hidden. It can be convenient to compare with other files on the desktop.

  • Running PRI Finder
  • If you want to mine informations of RNA-RNA interaction, please choose RNA button at right side corner, on the contrary, please choose Protein button.

  • Please choose a suitable input mode in the five modes.

  • Please choose a suitable threshold to mine informations.

  • Please choose a mode to show the result as you want.

  • Please choose a mode to handle the data. If you choose the mode of handling after downloading, please choose the cache directory to store the data.

  • Please click the start button, and then you can check the Massage to know the processing anytime.

  • You can export results when the processing is finished.

  • If you want to stop the processing, please click the stop button.

  • If you want to run PRI Finder again, please click the reset button to clear the Result window.

  • FAQ

  • PRI Finder Algorithm
  • Why does PRI Finder use the N-gram statistical language model?

  • N-gram statistical language model has the features of high flexibility and less manual intervention and so on. At the same time, it's positive corpus can be expanded anytime.

  • Why does PRI Finder use the Bigram statistical language model rather than the Trigram?

  • Bigram and Trigram can bring similar performances, but Trigram will use much longer time than Bigram to statistic the same data.

  • Where does our positive corpus come from?

  • In PRI Finder, the positive corpus was download from RAID database, which is the only one database to describe RNA-RNA interaction collecting by manual reading literatures.

  • How to deal with the sentence score when some bigram term does not appear in our positive corpus?

  • If some bigram term does not appear in our corpus, the probability of this bigram term will be equal to zero, and the score of this sentence will be equal to zero also. To avoid this result, we quoted katz smooth method ensuring this bigram term can be given a very small value instead of zero.

  • What's the named entity recognition?

  • When some RNA or protein appears in one sentence, we use "RNA-name" or "Protein-name" instead of original RNA or protein by Genia corpus.

  • How does PRI Finder mine the interations of RNA-RNA and Protein-Protein?

  • We set the threshold expressed by number 1 to 20. Each number is a certain degree of FDR. Larger the number, higher the false positive probability.

  • Why does the threshold 5 is the most modest?

  • The threshold five has the highest F1-score, which is expressed by both property of the precision and the recall.

  • Does the length of one sentence make a certain degree of influence?

  • Yes, it does. To avoid the influence of the length of one sentence, we use the geometric mean processing to take away this influence.

  • How to deal with the punctuation of one sentence?

  • When the data was input the software, it will be processed by punctuate treatment, and take away all the punctuations of each sentence.

  • How to deal with the capital letters and lower case letters?

  • All the letters will be changed into lower letters when the data is processed by punctuate treatment.

  • How to deal with different tenses of the same word?

  • All the words will be changed into original words, taking away all the tenses, when the data is processed by punctuate treatment.

  • How to optimize the effect of Punctuate treatment of the abstracts or the full texts?

  • When the data is processed by the punctuate treatment, to promote the effect, the software will set some conditions. Such as, when meeting "(", it will disconnect at the punctuation after occurring ")". When meeting ".", it will search the words after punctuation ".". If the letters after the punctuation "." are less than 30 letters, these words will be collected with the last sentence.

  • PRI Finder Data Files
  • What's the input mode of Single Pubmed ID?

  • It needs to input one pubmed ID.

  • What's the input mode of Pubmed ID List?

  • It needs to input one txt format file, in which there are series of pubmed ID.

  • What's the input mode of Single Text?

  • It needs to input one txt format file, which is an abstract or a full text of one literature.

  • What's the input mode of Text Directory?

  • It needs to input one document folder, in which there are series of txt format files.

  • What's the "PRI Finder the search"?

  • PRI Finder provides a platform to search literatures online. When choose "Download Pubmed ID List from NCBI", you can search literatures with key words and time range on NCBI.

  • How to export the results of input data files? What's the format of the exported file?

  • When PRI Finder run the data over, you can click Browse button to choose a directory, and export the result file. The result file is expressed by sentences in Excel with pubmed ID and score and flag of each sentence. The flag is the label of one sentence, which presents whether the score of the sentence is in the range of the threshold or not.

  • How to know the status of the processing of the input data files?

  • PRI Finder provides the Massage window to show the status of the process.

  • RAID
  • What's RAID?

  • RAID is the only one database to describe RNA-RNA interaction, whose data was collected by manual reading literatures.

  • Why does we put RAID in our web site?

  • First, the positive corpus of PRI Finder was downloaded from RAID database; Second, it can provides a convenient way to help you get analysis of RNA-RNA interaction.

  • How to use RAID?

  • When you click the RAID button, you will get into RAID database web site, and you will see the Help button. It will teach you how to use RAID in detail.

  • PRI Finder Software
  • What operating system do we need to run PRI Finder?

  • You should choose one computer with Windows operating system.

  • What's the meaning of each colour of sentence in Result?

  • When you run PRI Finder, the window of Result will show the sentence with different colors. The red color means this sentence has the score in the range of threshold, which means the content of this sentence may describe RNA-RNA interaction or Protein-Protein interaction. The yellow color means the score of this sentence is out of the range of the threshold, and greater than zero. The background color, green, means the score of this sentence is equal to zero.

    Software License

    For the software, we provide the download of source code on windows system, which was compiled in C++ and QT graphics library. However, if you are interested in the PRI Finder source code for your own use, while, your computer operating system is not windows, we ask you to please contact us for source code of PRI Finder on other operating systems.

    Software Acknowledgements

    PRI Finder implementations use many software tools and libraries, several of which are open source or freely available. Our thanks to their develops.

  • C++

  • QT graphics library

  • Mobanwang

  • Excel

  • Txt

  • Photoshop

  • Ue

  • For Broad Users

    The PRI Finder software was compiled on windows system. In order to adapt other operating systems, we provide following two versions of PRI Finder.

    PRI Finder – Windows Download

    Hello people! Welcome to our PRI Finder!

    The Interface of the PRI Finder Software:

    Click here to download the PRI Finder software (Only for windows)

    Hello people! Welcome to our PRI Finder!

    Contact:

    PRI Finder is developed at the Biomedical engineering lab  of Harbin Medical University. For detailed information on the PRI Finder method see the About page. If you have questions, comments or suggestions, we'd like to hear them. Contact us at: 854767460@qq.com

    We are here:


    Hello people! Welcome to our PRI Finder!

    这里是Contact内容















    about page content