I am working in the domain of complex disease medical research, with specialization across cross-disciplines of bioinformatics, biostatistics, computational genetics, high-performance computing (HPC) and AI/ML. I am primarily associated with the Swedish medical university, Karolinska Institutet, affiliated with the Department of Clinical Science and Education at the Södersjukhuset (KI SÖS), one of the largest hospitals in Stockholm. It involves part-time consulting for Prof. Erik Melén who is the HOD and is involved with multiple research projects, into understanding the complex interplay between genetic and environmental factors involved in lung growth, respiratory and allergic diseases, and based upon a birth cohort called BAMSE. My Karolinska profile is here. Additionally, I also consult for some start-ups in medical research, and am involved in developing Indo-European collaborative research projects in complex diseases as well as technology convergence in the genomics domain. 

Previously, I was working with the Diabetes Research group, led by Prof. Mark McCarthy at the Centre for Human Genetics, University of Oxford (from 2006-2014) and the Genetic Epidemiology group, led by Prof. Nicole Probst-Hensch at the Swiss Tropical and Public Health Institute, University of Basel in Switzerland (from 2010-2018).

Research Summary:

• OMICs data and multi-omics analytics involving combination of GWAS and imputed genetic variants, sequencing (genomic and exomes) biomarkers, methylation/epigenomics data, transcriptomics, expression arrays, proteomics and other related data. 

• Application of methodologies in artificial intelligence/machine learning for longitudinal trajectory modelling of epidemiology data.

• Complex diseases like asthma, type-2 diabetes, pulmonary function traits like lung function, COPD, eczema, rhinitis, obesity, BMI and related phenotypes. 

• GWAS analysis in consortium projects like DIAGRAM, ENGAGE, EGG, EAGLE, SpiroMeta, MAGIC, GoDMC, etc. 

• EWAS analysis in consortium projects with PACE and MeDALL

• Discovery, genotyping and association evaluation of variants in projects like GoT2D, GABRIEL and SUMMIT project (developing SOPs and pipelines for discovery/analysis on HPC servers and super-computing clusters).

• Evaluations of methylation chip data using various normalisation methods

• Annotation challenges with genetics and epigenetics datasets.

Expertise:

• Covering diverse range of platforms including low-coverage sequencing, exome-sequencing, dense-array genotyping on various chips from Illumina & Affy, methylation arrays from Illumina, custom i-Select chips like Metabochip, HumanCoreExome,  etc.

• Multivariate statistics like logistic regression, principal components analysis, etc. to analyse various cohort datasets.

• Gene and environment interactions, rare variants and fine mapping analysis.

• Building and incorporating large QC and bioinformatics-pipelines for various types of -omics datasets like genotyped/imputed SNPs, methylation CpGs, RNA-seq, proteomics, expression, etc.

• Managing Data centre; high-performance computing resources.

• Large scale data storage, handling, formatting and visualisation. 

• Biostatistical analysis in pre-clinical data. 

• Bioinformatics education & training.

In Stockholm at Karolinska, the work is based upon various OMICs analytics projects around GWAS/imputed/methylation/gene-environmental interactions and data integration around these technologies in the BAMSE cohort and other collaborative studies world-wide.

My work at SwissTPH Basel, was focussed on the genetics of complex diseases and environmental interactions. The Genetic Epidemiology group in Basel investigates aetiology of underlying chronic diseases, and is involved with large cohort studies, such as the internationally renowned Swiss SAPALDIA study (ongoing for 20 years), and other collaborations world-wide. 

My work at Oxford (with McCarthy group) was focussed on the genetics of Type 2 Diabetes (T2D) and related phenotypes on GoT2D, SUMMIT and other projects. GoT2D project looked into the genetic architecture of type 2 diabetes (T2D) using a combination of low-coverage sequencing data, exome data and 2.5M Omni chip array from Illumina. The projects are spread across various international consortia like DIAGRAM (T2D genetics), EU-funded ENGAGE consortium (www.euengage.org), GIANT (anthropometric traits), MAGIC (continuous glycaemic traits), etc.

Earlier, I was in the Department of Cardiovascular Medicine, working with Prof. Martin Farrall on a EU Framework 6 grant called GABRIEL, co-ordinated by Prof. Bill Cookson. The project aimed at investigating the genetic and environmental causes of asthma in the European Community. The first phase of genetic analysis involved GWAS (Genome-Wide Association Studies) across 994 cases and 1243 controls (from 2 cohorts), and led to the discovery of link between ORMDL3 and childhood asthma. The second phase involved 22 cohorts for GWAS across approximately 25000 cases and controls. For this, the genotyping was done on Illumina 610-Quad BeadChips at CNG, Paris.

I've commissioned and manage various high performance computing servers like DellTM  PowerEdgeTM  R930 (four socket, with upto 24 cores per processor, Intel® Xeon® processor E7-xxxx product family, 512 GB RAM, extendable up to 6TB DDR4 memory on 96 DIMMs), DellTM  PowerEdgeTM  R820 (four processors, with up to ten cores per processor, Intel® Xeon® processor E5-xxxx product family and hyper-dense, xx-DIMM memory, upto 1Tb RAM), DellTM   PowerEdgeTM  R905 (four Six-Core AMD OpteronTM8000 Series processors, 128 GB RAM) and Sun Fire V40z server (four dual-core AMD OpteronTM  64-bit CPUs, 16 GB RAM) with Red Hat Enterprise Linux, Centos, Debian and Ubuntu distributions. Also attached are various capacities of "Storage Array" solutions for upto 140TB space, involving various RAID configurations for data storage.

Previously, I have worked in the industry for around 8 yrs; in the UK as Bioinformatician / IT analyst with companies like Unilever (on the pea genome project, at their Sharnbrook corporate research centre) & Volkswagen (VW group national learning centre in Milton Keynes); in India with some promising start-ups like Onicra, LinkEdge Technologies, Career Launcher and Netaquila Solutions; providing IT consultancy, project management, business development and client account management, business process automation, ERP - Enterprise Resource Planning, database development, data migration and ware-housing, b2b.

Technology Overview

•  Various open-source bioinformatics tools like Plink/Plinkseq, GATK suite, vcftools and other sequencing softwares, Bioconductor R package, Eigenstrat, IMPUTE, MACH/minimac, ShapeIT, PHASE, SNPtest, platypus, EPACTS, SnpEff, Picard, etc.

•  Tools and practices – Stata, R, SAS, Matlab, SPSS, Citrix, Visio, MS-Project, MS-Office, data mining, data management, scientific programming, algorithm development, AI and ML, big-data visualisation. 

•  Language - HTML, Perl/python, BASH/AWK, CGI, Java/J2EE, various scripting, VB, .NET, C/C++, UML, XML

•  Linux admin - RedHat, Centos, Ubuntu, networking (TCP/IP, DNS, FTP), High performance computing/cluster pipelining of big data analysis, AWS, Server management tools like DELL's "iDRAC", etc. 

•  MS Access 97/ 2000, MS SQL Server 2005, MySQL 5.0, Oracle9.x, FoxBASE, FoxPRO

•  Server tools – RHEL 5/6/7.x, 64-bit Ubuntu/Debian Linux, WinNT and 2000, TSM, Backup Exec 8.x, user set-ups, access rights, various propriety and open-source application setup, data security, MS exchange, SSH, RAS, Tight VNC Viewer, excellent understanding of web-architecture.

Publications

Click here for a detailed listing in academia.

Contact Details:

Ashish Kumar

Department of Clinical Sciences and Education | Södersjukhuset (SÖS) | Karolinska Institutet.

Postal address : 

Sachsska, Sjukhusbacken 10, 

118 83 Stockholm, Sweden.

Email – ashish.kumar<at>ki.se