Predicting inframe indels in human pathogenicity

Since ancient times, human beings have never stopped fighting various diseases. Recently, researchers study the occurrence, development and treatment of diseases at the genetic level. The discrimination of pathogenic from benign mutations have been a hot topic in the field of pathobiology. However, there are few methods focusing on inframe indels variants. Here, we present a computational method for predicting pathogenicity impacted by inframe insertion/deletion variations.

PredinID - Prediction for inframe indels

PredinID is a web application developed to discriminate human inframe indels as pathogenic or benign. More specifically, we developed a Graph Convolutional Network (GCN) classifier based on edge sampling with a series of biological features.

Email Address

Please input or paste

Please provide transcript, position information in cDNA and gDNA. Data must be space-separated (at most 50 variants).

***  Example: ENST00000378536 c.279_287delGTCCGACCG g.2160484_2160492delGTCCGACCG  ***


Zhenyu Yue, Xiang Ying and Guojun Chen. PredinID: predicting pathogenic inframe indels in human through graph convolution neural network with graph sampling technique.


PredinID can be complied and used on Linux. As well it requires Python and R to work properly. As input, the PredinID may desire the file from conservation scoring (PhyloP) to help generate features about evolutionary conservation scoring. To install and run PredinID Indel, you can download the dataset and code from here.

Contact us

We welcome any suggestions on how to improve our service, please do not hesitate to contact us with feedback.

School of Information and Computer, Anhui Agricultural University, Hefei, Anhui 230036, China