Chronic non-communicable diseases （NCDs） is a sick- state in the complex system of human body. Healthsystems engineering approach can be used to recover health state by many different ways of system adjustments, including physical exercises, nutrition and psychological intervention, and drug therapies. To evaluate and assess the effects of these adjustments, many parameters of the system states have to be monitored （such as molecular high- throughput technologies, physiological, image, etc.）, and estimated at the different levels （such as molecular, cellular, tissue, organ, and system levels） and on the different dimensions （such as metabolic, immune, neural, etc.） for human system. Huge data have been produced during the whole process, and health systems engineering approach will model health and sick state in the human system by detecting and analyzing these multi-dimension and multi- level big-data in order to find personal suitable adjustment method.
Although the diploid nature has been observed for over 50 years, phasing the diploid is still a laborious task. The speed and throughput of next generation sequencing have largely increased in the past decades. However, the short read-length remains one of the biggest challenges of haplotype analysis. For instance, reads as short as 150 bp span no more than one variant in most cases. Numerous experimental technologies have been developed to overcome this challenge. Distance, complexity and accuracy of the linkages obtained are the main factors to evaluate the efficiency of whole genome haplotyping methods. Here, we review these experimental technologies, evaluating their efficiency in linkages obtaining and system complexity. The technologies are organized into four categories based on its strategy： （i） chromosomes separation, （ii） dilution pools, （iii） crosslinking and proximity ligation, （ix） long-read technologies. Within each category, several subsections are listed to classify each technology. Innovative experimental strategies are expected to have high-quality performance, low cost and be labor-saving, which will be largely desired in the future.
High-throughput technologies were employed over the past decade to study the expression profiles of cells and tissues.There are large collections of accumulated data from public databases and numerous research articles were published on these data.In the current study,we performed meta-analysis on the gene expression data from human liver and kidney tissues produced from five different technologies:EST,SAGE,MPSS,microarray,and RNA-Seq.We found RNA-Seq was the most sensitive in the number of genes it detected while SAGE and MPSS were the least sensitive.For the genes detected by all the platforms,there were generally good correlations to the measured expression levels of corresponding genes.We further compared detected genes to liver/kidney proteomics data from the Human Protein Atlas,and found 960 of the 8764 genes only detected by RNA-Seq were validated by proteomics results.In conclusion,RNA-Seq is a more sensitive and consistent technology compared to the other four high-throughput platforms,though their results are in general agreement.Average coverage was determined to be the preferred measurement to represent gene expression levels by RNA-Seq data and will be used in future works.
