Welcome to the live demo of the website! During the demo, the controls of the application will be disabled. You can stop the demo any time by clicking 'EXIT' button or reloading the page. The demo will highlight buttons and fields in red at each step to draw your attention to them. The guiders are draggable, and thus you can move them if the view of any element of interest is obscured.

This demonstration will guide you through the process of variant prioritisation using OVA. For the purposes of this demonstration, we will prioritise a VCF file of a patient with Primary Ciliary Dyskinesia. Click 'Next' to Proceed.

We can define a patient phenotype either as a set of terms or as one or more diseases. Here, we will search by disease by selecting 'OMIM disease' radio button and typing the disease name into the field.

We select relevant phenotypes/diseases from the search results and press 'Proceed'.

OVA can prioritise gene lists ( e.g. list of top differentially expressed genes), genomic regions ( e.g. GWAS studies) or variants called from next generation sequencing data. Radio buttons can be used to select relevant input types.

Here, we will prioritise a VCF file of the patient. Various filtering options are available which can greatly aid prioritisation. In this example, we are searching for an autosomal recessive gene, so we will keep only homozygous or compound heterozygous variants. We will also filter out synonymous, intronic and UTR variants.

We are also interested in filtering out small in-frame indels which are common in the human genome and are unlikely to be pathogenic. We can set the maximum size of the indels to discard by clicking the link underneath the box

Similarly, we can define the size of splice sites. Intronic variants which fall within this user-defined region will not be filtered unless 'Splice Sites' is selected. We will accept default setting for this demo.

Finally, there's a possibility to further narrow down the search by defining genomic regions. Our patient comes from a consanguinous family and we have identified a number of autozygous intervals. We can enter these either manually or by uploading a text file.

We click 'Upload' to upload and filter our VCF file based on selected criteria. This may take a moment. The tooltip will report how many genes and variants remain after the filtering is complete. We click 'Proceed'.

Our algorithm aims to select appriopriate Gene Ontology terms to your query phenotype. However, sometimes this cannot replace human knowledge. This is a review step which allows the user to remove irrelevant terms or add extra functions they know are important to the query phenotype. Once done, we press 'Proceed'.

Prioritisation may take a moment, depending on query size and current server loads. Once done, the top 200 results are presented in a table. Full results can be downloaded in a text file.

Please wait for prioritisation to finish to proceed to the next step.

The results table is fully sortable by score types. We know that cilium is an important in PCD, thus we may be interested in seeing which genes have the highest cellular component score. We click 'CC Score, Human' to sort the table.

We can have a look at the annotations of each individual gene by clicking 'View' button in the 'View GO Annotations column'. Terms which are similar to your query are larger, while less similar ones are smaller.

We can view individual variants which passed the filtering criteria for each gene by clicking 'View' button in the 'View Variants' column. LRRC6 gene harbours 2 missense mutations as well as a single base pair deletion which will cause a frame shift. Taken together with gene functional annotations, these mutations indicade a good candidate gene that is worth following up.

You have reached the end of the demo. We hope it has been informative. Any further questions can be directed via 'Contact' page at the bottom.

OVA: Ontology Variant Analysis Tool

Gene Selection