Experimental methods and the big data sets
- 1. Improved Medical Education in Basic Sciences for Better Medical Practicing ImproveMEd Systems biology for medicine II. Experimental methods and the big data sets
- 2. Experiments in systems biology do not need to be omic-scale to satisfy criteria of system biology! Consider experiments focused on subsystems like e.g. tracking energy metabolism relevant mRNA under different feeding regime, or at multiple time points from onset of feeding. What system biology differ from common scientific research are: 1. Quantitative data – where, how much and how fast (dynamic, time-dependance) will an entity change 2. Computational models - simulations based on precise quantification and timing Still need positive and negativ controls and at least 3 replicates! 2. Hypothesis driven studies that follow targeted subset of molecules (or targeted organelle) – small scale systems biology. 1. Hypothesis generating studies!
- 3. Molecules: • DNAs microarrays & sequencing based technologies • RNAs mRNA sequencing • Proteins Mass spectrometry-based proteomics • Lipids liquid chromatography & Mass spectrometry • Metabolites liquid chromatography Mass spectrometry
- 4. Population Average Techniques • Population of cells or tissue sample ≈ 1 million cells or more • Average over many cells Single cell techniques • Single cell sample • Cell-to-cell variability HepaRG stable cell lineLiver tissue Hepatocyte spheroids
- 5. Microarrays – differential expression • Dominating technique in 2000s • Predominantly used for measuring transcript levels • Other uses: genotyping, DNA mapping (copy number variation), DNA methylation • Microarray consists of spots printed with different oligonucleotide • Hybridization between fluorescently labeled sample (probes) and printed oligonucletides • GEO database https://www.ncbi.nlm.nih.gov/geo/info/qqtutorial.ht ml
- 6. Array CGH comparative genomic hybridization • molecular cytogenetic method/ cariotypisation • CNV relativ to test sample • Based on assumtion that 2 samples from closely related individuals differ (healthy and sick) in gain or loss of chromosome or a chromosomal region • Large scale analysis of tumor-specific DNA unbalanced rearangements with resolutio of 5-10 megabases • OMIM database https://www.ncbi.nlm.nih.gov/omim
- 7. Microarrays Methylation Assay - epigenome • Epigenetic regulation of gene expression important in development, genetic imprinting, tumorogenesis… • Genome wide CpG methylation level at 30 000 till 500 000 CpG loci covering whole genome • The first step is bisulfite conversion of samples – unmethylated sites convert C into U, while methylated lose methylation • Converted DNA is further amplified (and U is exchanged for T) • Fragmented and denatured oligonucleotides are hybridized with two types of allel specific beads for each locus • The final nucleotide of anneiled allel specific oligonucleotide is futher extended by labeled dDNT • Software calculate relative fluorecence of each locus and grade as 0, 0.5 or 1 (homozygous unmethylated, heterozygous, or homozygous methylated) • ENCODE https://genome.ucsc.edu/encode/dataMatrix/encodeChipMatrixHuman.html
- 8. Sequencing-based technologies • Starts with DNA or RNA isolation followed by either DNA fragmentation or cDNA synthesis • Next step is amplification (clone fragments into vector, trasform bacteria with vectors and amplify) • Parallel sequencing of many short DNA pieces • Assemble of contiguous fragments
- 9. Whole-genome sequencing (WGS) vs. whole-exome sequencing (WES) • WGS attempts to sequence entire genome. Some sequences are technically challanging to sequence (telomeres, centromeres, high CG content or repeating loci) and they been missed by common sequencing platforms what results in 95-98% of genome coverage, but on uniformed way. • WES sequence just exomes (coding sequences) or 2% of genome, by each exomes is sequences 30-100x (high depth). DNA-RNA hybridization is used for selection of coding regions what causes over- representation of so called ’hot-spots’ and unde- representation of missed variants. Quicker, chipper and easier data analysis, but low overall coverage. uniform bias
- 10. Exome sequencing use an enrichment step to select targeted DNA • Mendelian disorders often disrupt protein-coding regions • Exom is a good source of rare disease variants • Microarreys can be used for fragment isolation • WES gives high depth of sequencing • Bamshat et al. (2011) Nature Review Genetics
- 11. Cost per genome won the race and WGS becomes a preferable method, especially for tumor-specific rearrangements Evolution of sequencing methodes 1. The first generation Sanger sequencing; chain termination technology – ddNTPs used to terminate chain synthesis further separated by capillary electrophoreses (one strand at a time, slow, accurate, expensive) 2. The second (next) generation sequencing – sequencing by synthesis – introduction of nano-technologie - parallel sequencing and no need for separation step – limits in read length 3. The third generation sequencing – immobilized polymerase + fluorescent dNTP + excelent optics – detection of base incorporation and base modification
- 12. Data coming out WGS or WES need verification done over a large population! (more than 1000 individuals) The database of Genotypes and Phenotypes (dbGaP) https://www.ncbi.nlm.nih.gov/gap Genome-wide association studies Foo et al. (2012) Nature Review Neurology
- 13. RNA-Seq transcriptome • Parallel sequencing of mRNA, rRNA, miRNA… • Gene expression profile • Quantitative method ideal for systems biology experiments • Identification of alternative splicing variants and new transcript variants • GEO database https://www.ncbi.nlm.nih.gov/geo/info/qqt utorial.html • Target Scan databas (miRNA) http://www.targetscan.org/vert_71/ Wang et al.(2009) Nature Reviews Genetics
- 14. RNA-seq become a dominate transcriptome methode and replaced microarray Wang et al.(2009) Nature Reviews Genetics
- 15. ChIP-seq Chromatin Immunoprecipitation Sequencing • Combines precipitation of transcription factors or other DNA binding proteins (using specific antibodies) or histone modifications and deep sequencing of co- immunoprecipitated DNA • Discovers regulatory sequences, promoters, enhancers, silencers, spacers… • Functional organization of genome
- 16. Western Blot was the first method toward quantification and identification of more than one protein • Semi- quantitative because of non-linear kinetics behind enzyme –labeling of secondary antibodies and substrate reaction • Also has non-linear kinetics in the case of chemiluminescence reaction or X-ray film exposure • LICOR is infrared fluorescence that gives little background and fluorescent signal is linearly proportional to amount of antibody • https://www.licor.com/bio/applications/quan titative_western_blots/
- 17. Forward and Reverse Phase Protein Array (FPPA &RPPA) • Attempt to transforme low throughput Western blot methode (8-16 lanes for samples) into high throughput method • In forward version one antibodie is spotted on slide and mane samples are probed for presence of epitop • In reverse version many antibodies (with high specificity) are spoted on slide and one sample is probed for all antibodies. • Requires high specificity antibodies, expensive.
- 18. Mass Spectrometry proteomics, lipidomics, metabolomics • Proteomics often use tandem mass spectrometry (two mass specs done in paralle) • Quantitative because measures total level of proteins • Gives detailes about post-translational modifications • If combinened with immunoprecipitation gives information about protein interactions • Involves many steps: separation, digestion, enrichment, repeated separation, ionization, mass filtering (MS1), fragmentation and mass analysis (MS2), identification, quantification • Many variations
- 19. Mass Spectrometry proteomics, lipidomics, metabolomics The final step is based on the large data base of known peptides and bioinformatics search for matches. Different version of MS and different data bases are used for lipidomics and metabolomics. UniProtKB database https://web.expasy.org/docs/swiss- prot_guideline.html
- 20. Proteomics analysis decoded differences between MAPK pathway in normal and tumor cell. Choudhary & Mann (2010) Nature Reviews Molecular and Cell Biology
- 21. proteomics, lipidomics, metabolomics The final step is based on the large data base of known peptides and bioinformatics search for matches. Different version of MS and different data bases are used for lipidomics and metabolomics. UniProtKB database https://web.expasy.org/docs/swiss- prot_guideline.html
- 22. Liquid chromatography (LC) & LC/MS lipidomics, metabolomics Palermo et al. (2017) Analytica Chimica Acta Very often LC & MS are combined and used in sequence. Also, lipidomic & metabolomics could be done in sequence on the same samples. LC is used to divide sample in fractions, as an separation technique, while MS is used for identification of molecules.
- 23. Systems biology experiments collect the big data using a high throughput methods: • DNAs microarrays & sequencing based technologies (NGS) exome/genome • DNA + regulatory proteins ChIP-seq ReMap • RNAs RNA-seq transcriptome • Proteins MS proteome • Lipids LC/MS lipidom • Metabolites LC/MS metabolom