Kiran Kumar

Experience

• 

  PointCross

  Dec 2008 - now

  I involved in analysis of requirements, design and development of different components for XBIOM Insights. Developing precisely targeted therapies requires researchers to have direct access to a high volume and variety of data from clinical data of subjects from trials, their genomic profiles and molecular biomarkers from bio-samples. They need to refer to bioinformatics registries for genes, proteins or miRNA and their known mutations. Today, finding this data and preparing it into a consumable presentation for researchers is time consuming and requires considerable effort by data managers and programmers. Finding highly stratified cohorts that match the researcher's hypothesis of the genotype, phenotype and medical history is very expensive and takes many weeks to test just one hypothesis. Xbiom™ Insights for Clinical Biomarkers eliminates the need to look for, and prepare this data, by semantically integrating the data from disparate sources with automation in mapping and loading while minimizing IT dependency. XbiomTM empowers researchers through its simple, editable, and re-usable query masks and search facility to deliver immediate results. The platform supports translational analytics across clinical, pre-clinical, molecular and other biomarker data to human users and remote applications that access XbiomTM for its platform services. This is used to get research data from top pharma companies worldwide and import that data into Universal data modal which is globally agreed using SDE(Semantic Data Exchange) which we build for importing data from different sources. The platform covers the R&D needs for data integration, search, discovery and translational analytics with its components for clinical, biomarker, and nonclinical data. Show less I am involved in analysis of requirements and design and development of different components for NIMS (Non clinical information management system). ). This is used to get research data from FDA (food and Drug Authority) and import that data into Universal data modal which is globally agreed and called as SEND data modal using SDE(Study Data Exchange) which we build for importing data from different sources to database(Oracle, Postgrase or Sql server) . After getting data into universal data model we created different Taxonomies like parameters, Sample types, category, species, animal status which is used for search and data viewing, and used different statistical methods to calculate statistical values like t-test, Man Witney test, Quintiles, Outliers, Mean, Median, Standard deviation, and range and index these data along with source data in SOLR which we used to search and get data (this component we used for getting better performance while searching) and build different Toxicology visions and generated different . And in our application user can annotate the data.This product is mainly designed for FDA so the users are reviewers which are more interested in playing with data and find some similarities, and interesting trends. And for reviewers we have to give options to make their favourites like studies and show those for them when they enter in to the system. In this product we provided few more functionalities like clubbing multiple groups and generating data and statistical data in run time, showing some consolidated views, Historical control reference ranges, and exporting the viewed data into different formats like Microsoft excel, Microsoft word document, XML, XSD, and Prism formats. And allowed users to view many to many relations like within study how many compounds are there and their property as well as one compound in used in how many studies and those details. Show less I am involved in design and development of SDIS (Study data integration and search system). This is used to get research data from different Non-clinical research companies (who are facing problems in organising data and searching required data) and import that data into Universal data modal which is globally agreed and called as SEND data modal. After getting data into universal data model we use different statistical methods to calculate statistical values like t-test, Man Witney test, Quintiles, Outliers, Mean, Median, Standard deviation, and range and index these data along with source data in SOLR which we used to search and get data (this component we used for getting better performance while searching) and build different Toxic visions and generated different charts like box and whisker, X-bar, Bar and Line along with statistical clustering and data viewing charts like Heat map, and Z-transformation. And in our application user can annotate the data and while switching between views user can view data in his own way and which is in more drill down way like from study to groups and from groups to subject and within subjects and groups user can view different pre-clinical domains, parameters, category, panels and sample types data part by part and by combining all. Show less

  • Technical Lead

    Feb 2011 - now

  • Technical lead

    Dec 2008 - now

  • software Enginnering

    Dec 2008 - Feb 2011

  • Software engineer

    Dec 2008 - Mar 2010