Canadian Bioinformatics Workshops

1
Canadian Bioinformatics Workshops

• www.bioinformatics.ca

2
2
Module Title of Module
3
Module 1 Introduction to Metabolomics

• David Wishart
• Informatics and Statistics for Metabolomics
• June 15-16, 2015

4
Learning Objectives

• To define metabolomics and the size of the
  metabolome(s)
• To appreciate the importance and potential
  applications of metabolomics
• To understand the operational principles of key
  metabolomics technologies (LC, GC, MS and NMR)
• To understand the difference between targeted and
  untargeted metabolomics

5
Schedule
6
The Pyramid of Life
Metabolomics Proteomics Genomics
Metabolome
Environmental Influence
Physiological Influence
Proteome
Genome
7
What is Metabolomics?

• Genomics - A field of life science research that
  uses High Throughput (HT) technologies to
  identify and/or characterize all the genes in a
  given cell, tissue or organism (i.e. the genome).
• Metabolomics - A field of life science research
  that uses High Throughput (HT) technologies to
  identify and/or characterize all the small
  molecules or metabolites in a given cell, tissue
  or organism (i.e. the metabolome).

8
What is a Metabolite?

• Any organic molecule detectable in the body with
  a MW lt 1500 Da
• Includes peptides, oligonucleotides, sugars,
  nucelosides, organic acids, ketones, aldehydes,
  amines, amino acids, lipids, steroids, alkaloids,
  foods, food additives, toxins, pollutants, drugs
  and drug metabolites
• Includes human microbial products
• Concentration gt detectable (1 pM)

9
What is a Metabolome?

• The complete collection of small molecule
  metabolites in a cell, organ, tissue or organism
• Includes endogenous and exogenous molecules as
  well as transient or even theoretical molecules
• Defined by the detection technology
• Metabolome size is always ill-defined

10
Different Metabolomes
All Mammals All Microbes All Plants
60,000 Chemicals
100,000 Chemicals
300,000 Chemicals
The Pyramid of Life
11
Human Metabolomes (2015)
3670 (T3DB) 1240 (DrugBank) 28500
(FooDB) 1550 (DrugBank) 19700 (HMDB)
M mM ?M
nM pM
fM
12
Theoretical Human Metabolomes
100,000 (Lipidome) 10,000 (Drug
metabolome) 100,000 (Food metabolome) 10,000
(Secondome)
M mM ?M
nM pM
fM
13
Why is Metabolomics Important?

14
Small Molecules Count

• gt95 of all diagnostic clinical assays test for
  small molecules
• 89 of all known drugs are small molecules
• 50 of all drugs are derived from pre-existing
  metabolites
• 30 of identified genetic disorders involve
  diseases of small molecule metabolism
• Small molecules serve as cofactors and signaling
  molecules to 1000s of proteins

15
Metabolites Are the Canaries of the Genome
A single base change can lead to a 10,000X change
in metabolite levels
16
Metabolomics is More Time Sensitive Than Other
Omics
Response
Metabolomics
Response
Proteomics
Response
Genomics
Time
17
Metabolism is Understood
18
The Metabolome is Connected to all other Omes
Meta bolome
Proteome
Genome
The Pyramid of Life
19
The Metabolome is Connected to All Other Omes

• Small molecules (i.e. AMP, CMP, GMP, TMP) are the
  primary constituents of the genome
  transcriptome
• Small molecules (i.e. the 20 amino acids) are the
  primary constituents of the proteome
• Small molecules (i.e. lipids) give cells their
  shape, form, integrity and structure
• Small molecules (sugars, lipids, AAs, ATP) are
  the source of all cellular energy
• Small molecules serve as cofactors and signaling
  molecules for both the proteome and the genome
• The genome proteome largely evolved to catalyze
  the chemistry of small molecules

20
Metabolomics Enables Systems Biology
Bioinformatics
Meta bolomics
Cheminformatics
Systems Biology
Proteomics
Genomics
21
Metabolomics Applications

• Genetic Disease Tests
• Nutritional Analysis
• Clinical Blood Analysis
• Clinical Urinalysis
• Cholesterol Testing
• Drug Compliance
• Transplant Monitoring
• MRS and CS imaging

• Toxicology Testing
• Clinical Trial Testing
• Fermentation Monitoring
• Food Beverage Tests
• Nutraceutical Analysis
• Drug Phenotyping
• Water Quality Testing
• Petrochemical Analysis

22
Metabolomics Methods

23
Metabolomics Workflow
Biological or Tissue Samples
Extraction
Biofluids or Extracts
Chemical Analysis
Data Analysis
24
Comparing Omics Coverage
Metabolomics Proteomics Genomics
200 Chemicals
Completeness
5000 Proteins
22,000 Genes
The Pyramid of Life
25
Why Metabolomics is Difficult
Metabolomics Proteomics Genomics
2x105 Chemicals
Chemical Diversity
20 Amino acids
4 Bases
The Pyramid of Life
26
Metabolomics Technologies

• UPLC, HPLC
• CE/microfluidics
• LC-MS
• FT-MS
• QqQ-MS
• NMR spectroscopy
• X-ray crystallography
• GC-MS
• FTIR

27
Chromatography
28
Chromatography

• The separation of components in a mixture that
  involves passing the mixture dissolved in a
  "mobile phase" through a stationary phase, which
  separates the analyte to be measured from other
  molecules in the mixture based on differential
  partitioning between the mobile and stationary
  phases
• Column, thin layer, liquid, gas, affinity, ion
  exchange, size exclusion, reverse phase, normal
  phase, gravity, high pressure

29
High Pressure (Performance) Liquid Chromatography
- HPLC

• Developed in 1970s
• Uses high pressures (6000 psi) and smaller (5
  mm), pressure-stable particles
• Allows compounds to be detected at ppt (parts per
  trillion) level
• Allows separation of many types of polar and
  nonpolar compounds

30
HPLC Modalities

• Reversed phase for separation of non-polar
  molecules (non-polar stationary phase, polar
  mobile phase)
• Normal phase for separation of non-polar
  molecules (polar stationary phase,
  non-polar/organic mobile phase)
• HILIC hydrophilic interaction liquid
  chromatography for separation of polar molecules
  (polar stationary phase, mixed polar/nonpolar
  mobile phase)

31
HPLC Columns
32
Reverse Phase Column
33
HPLC Separation Efficiency
34
HPLC Schematic
35
Gradient HPLC Schematic
36
HPLC of a Biological Mixture
37
Gas Chromatography
38
Gas Chromatography

• Involves a sample being vaporized to a gas and
  injected into a column
• Sample is transported through the column by an
  inert gas mobile phase
• Column has a liquid or polymer stationary phase
  that is adsorbed to the surface of a metal tube
• Columns are 1.5-10 m in length and 2-4 mm in
  internal diameter
• Samples are usually derivatized with TMS to make
  them volatile

39
TMS Derivatization
40
Gas Chromatography
41
GC-Columns
Polysiloxane
42
Retention Time/Index

• Retention time (RT) is the time taken by an
  analyte to pass through a column
• RT is affected by compound, column (dimensions
  and stationary phase), flow rate, pressure,
  carrier, temp.
• Comparing RT from a standard sample to an unknown
  allows compound ID
• Retention index (RI) is the retention time
  normalized to the retention times of adjacently
  eluting n-alkanes

43
Compound Identification and Quantification
44
GC-MS Chromatogram of a Biological Mixture
45
Mass Spectrometry

• Analytical method to measure the molecular or
  atomic weight of samples

46
Typical Mass Spectrometer
47
MS Principles

• Different compounds can be uniquely identified by
  their mass

Butorphanol L-dopa Ethanol
CH3CH2OH
MW 327.1 MW 197.2 MW 46.1
48
Mass Spectrometry

• For small organic molecules the MW can be
  determined to within 1 ppm or 0.0001 which is
  sufficiently accurate to confirm the molecular
  formula from mass alone
• For large biomolecules the MW can be routinely
  determined within an accuracy of 0.002 (i.e.
  within 1 Da for a 40 kD protein)
• Recall 1 dalton 1 atomic mass unit (1 amu)

49
Different Types of MS

• GC-MS - Gas Chromatography MS
• separates volatile compounds in gas column and
  IDs by mass
• LC-MS - Liquid Chromatography MS
• separates delicate compounds in HPLC column and
  IDs by mass
• MS-MS - Tandem Mass Spectrometry
• separates compound fragments by magnetic or
  electric fields and IDs by mass fragment patterns

50
Masses in MS

• Monoisotopic mass is the mass determined using
  the masses of the most abundant isotopes
• Average mass is the abundance weighted mass of
  all isotopic components

51
Isotopic Distributions
1H 99.9 12C 98.9 35Cl
68.1 2H 0.02 13C 1.1 37Cl
31.9
52
Isotopic Distributions
1H 99.9 12C 98.9 35Cl
68.1 2H 0.02 13C 1.1 37Cl
31.9
100
32.1
6.6
2.1
0.06
0.00
m/z
53
Mass Spec Principles
Sample

_
Detector
Ionizer
Mass Analyzer
54
Typical Mass Spectrum
aspirin
55
Typical Mass Spectrum

• Characterized by sharp, narrow peaks
• X-axis position indicates the m/z ratio of a
  given ion (for singly charged ions this
  corresponds to the mass of the ion)
• Height of peak indicates the relative abundance
  of a given ion (not reliable for quantitation)
• Peak intensity indicates the ions ability to
  desorb or fly (some fly better than others)

56
Resolution Resolving Power

• Width of peak indicates the resolution of the MS
  instrument
• The better the resolution or resolving power, the
  better the instrument and the better the mass
  accuracy
• Resolving power is defined as
• M is the mass number of the observed mass (DM) is
  the difference between two masses that can be
  separated

57
Resolution in MS
58
Resolution in MS
Low resolution Instrument (Ion trap)
2847
High resolution Instrument (TOF)
59
Resolution/Resolving Power
MW(mono) 3482.7473 MW(ave) 3484 Blue DM/M
1000 Red DM/M 3000 Green DM/M 10000 Black
DM/M 30000
60
Mass Spectrometer Schematic
61
Different Ionization Methods

• Electron Ionization (EI - Hard method)
• Small molecules, 1-1000 Daltons, structure
• Chemical Ionization (CI Semi-hard)
• Small molecules, 1-1000 Daltons, simple spectra
• Electrospray Ionization (ESI - Soft)
• Small molecules, peptides, proteins, up to
  200,000 Daltons
• Matrix Assisted Laser Desorption (MALDI-Soft)
• Smallish molecules, peptides, proteins, DNA, up
  to 500 kD

62
(No Transcript)
63
Electron Impact Ionization

• Sample introduced into instrument by heating it
  until it evaporates
• Gas phase sample is bombarded with electrons
  coming from rhenium or tungsten filament (energy
  70 eV)
• Molecule is shattered into fragments (70 eV gtgt
  5 eV bonds)
• Fragments sent to mass analyzer
• Most commonly used in GC-MS

64
EI Fragmentation of CH3OH
CH3OH
CH3OH
CH3OH
CH2OH
H
CH3OH
CH3
OH
CHOH
H
CH2OH
65
Electron Impact MS of CH3OH
Molecular ion
EI Breaks up Molecules in Predictable Ways
66
Soft Ionization Methods
337 nm UV laser
Fluid (no salt)

_
Gold tip needle
cyano-hydroxy cinnamic acid
MALDI
ESI
67
Electrospray (Detail)
68
Electrospray (Detail)
69
Electrospray Ionization

• Sample dissolved in polar, volatile buffer (no
  salts) and pumped through a stainless steel
  capillary (70 - 150 mm) at a rate of 10-100
  mL/min
• Strong voltage (3-4 kV) applied at tip along with
  flow of nebulizing gas causes the sample to
  nebulize or aerosolize
• Aerosol is directed through regions of higher
  vacuum until droplets evaporate to near atomic
  size (still carrying charges)

70
Electrospray Ionization
5H2O/95CH3CN
95H2O/5CH3CN
100 V 1000 V 3000 V
71
Electrospray Ionization

• Can be modified to nanospray system with flow lt
  1 mL/min
• Very sensitive technique, requires less than a
  picomole of material
• Strongly affected by salts detergents
• Positive ion mode measures (M H) (add formic
  acid to solvent)
• Negative ion mode measures (M - H)- (add ammonia
  to solvent)

72
Mass Spectrometer Schematic
73
Different Types of Mass Analyzers

• Magnetic Sector Analyzer (MSA)
• High resolution, exact mass, original MA
• Quadrupole Analyzer (Q or Q)
• Low (1 amu) resolution, fast, cheap
• Time-of-Flight Analyzer (TOF)
• No upper m/z limit, high throughput
• Ion Cyclotron Resonance (FT-ICR)
• Highest resolution, exact mass, costly

74
MS Mass Accuracy

 Mass Accuracy
Type
 0.1 - 1 ppm 
 FT-ICR-MS
 0.5 - 1 ppm 
 Orbitrap
 1 - 2 ppm 
 Magnetic Sector 
 3 - 5 ppm 
 TOF-MS
 3 - 5 ppm 
 Q-TOF 
 3 - 5 ppm 
 Triple Quad 
 50-200 ppm
 Linear IonTrap
(10 ppm in Ultra-Zoom)
75
Mass Chromatograms

• Standard output from an LC-MS or GC-MS
  experiment
• X-axis is retention time, Y-axis is signal
  intensity
• Total Ion Current (TIC) chromatogram is summed
  intensity across the entire range of masses being
  detected at every point in the analysis
• Base Peak chromatogram (BPC) is like a TIC but
  displays only the most intense peak in each
  spectrum
• Extracted Ion chromatogram (EIC) contains one or
  more analytes extracted from the TIC or BPC

76
Mass Chromatograms of Biological Mixtures
Tomato Extract
Arabidopsis Extract
77
NMR Spectroscopy
78
Explaining NMR
79
Principles of NMR

• Measures nuclear magnetism or changes in nuclear
  magnetism in a molecule
• NMR spectroscopy measures the absorption of light
  (radio waves) due to changes in nuclear spin
  orientation
• NMR only occurs when a sample is in a strong
  magnetic field
• Different nuclei absorb at different energies
  (frequencies)

80
Protons (and other nucleons) Have Spin
Spin up Spin down
81
Each Spinning Proton is Like a Mini-Magnet
Spin up Spin down
82
Principles of NMR
N
N
hn
hn
S
S
Low Energy High Energy
83
Bigger Magnets are Better
Increasing magnetic field strength
low frequency high frequency
84
A Modern NMR Instrument
Radio Wave Transceiver
85
NMR Magnet
86
NMR Magnet Cross-Section
Sample Bore
Cryogens
Magnet Coil
Magnet Legs
Probe
87
An NMR Probe
88
NMR Sample Probe Coil
89
1H NMR Spectra Exhibit...

• Chemical Shifts (peaks at different frequencies
  or ppm values)
• Splitting Patterns (from spin coupling)
• Different Peak Intensities ( 1H)

90
Chemical Shifts

• Key to the utility of NMR in chemistry
• Different 1H in different molecules exhibit
  different absorption frequencies
• Each compound can be defined by a unique pattern
  of chemical shifts (a fingerprint)
• Chemical shifts are mostly affected by
  electronegativity of neighbouring atoms, bonds or
  groups

91
Characteristic Chemical Shifts
92
Assigning Simple NMR Spectra
TMS
93
Assigning Simple NMR Spectra
94
NMR Spectra Need Fixin
Before
After
Baseline correction
Water suppression
Referencing
Shimming
Phasing
95
NMR Spectra Need Fixin

• Chemical shift referencing (TMS, DSS)
• Calibrates/normalizes chemical shifts
• Shimming
• Fixes line shape to look Lorentzian
• Phasing
• Fixes line shape to look absorptive
• Water suppression/removal
• Removes large water signal
• Baseline correction
• Makes spectrum look flat not wobbly

96
NMR Spectrum of a Biological Mixture
97
Technology Sensitivity
Unknowns
4
LC-MS or DI-MS
3
GC-MS TOF
Metabolites or Features detected (Log10)
2
NMR
1
Knowns
GC-MS Quad
0
M mM ?M
nM pM
fM
Sensitivity or LDL
98
Comparison
NMR (with cold probe) GC-MS DI-MS
Techniques
Metabolites Water-soluble (amino acids, organic acids, sugars) mainly water-soluble (some hydrophobic) Mainly hydrophobic (some water-soluble)
Types of samples Biofluids, plant, bacterial, animal tissue extracts, Food Biofluids, plant, bacterial, animal tissue extracts, Food Mainly biofluids
Sample Volume 100 µL (min) 30-50 µL (min) 10 µL
99
Comparison
NMR GC-MS DI-MS
Sample prep time 30 -120 min/20 samples 30 -120 min/20 samples 3-4 h for 96 samples
Run time 10-90 min/sample 30-60 min/sample 7 min/sample
Data Analysis 30-60 min / sample 30-60 min / sample 1-2 h for 96 samples
Limit of Detection 5 µM 100 nM 5 nM
No. of metabolites 20-150 20-50 100-180
Overlapping Metabolites 10-15 10-15 10-15
Cross-checking 10-30 10-30 10-30
100
Whats Possible

• NMR-based metabolomics (50-200 metabolites
  identified/quantified, mM sensitivity)
• GC-MS based metabolomics (70-120 metabolites
  identified/quantified, ltmM sensitivity)
• DI-MS based metabolomics (180 metabolites
  identified/quantified, nM sensitivity)
• LC-MS based metabolomics (300-500 metabolites
  identified/quantified, nM sensitivity)
• Lipidomics (3000 lipids identified and
  semi-quantified, nM sensitivity)
• Specialty phytochemical, nutrient, drug and
  pesticide analysis (mostly HPLC, nM sensitivity)

101
2 Routes to Metabolomics
Quantitative (Targeted) Methods
Chemometric (Profiling) Methods
102
Profiling (Untargeted)
Data Reduction
Data Collection
Sample Prep
Metabolite Identification
103
Quantitative (Targeted)
Sample Prep
Biological Interpretation
Data Reduction
Metabolite Identification Quantification
104
From Spectra to Lists
105
From Lists to Pathways
106
From Pathways Lists to Models Biomarkers
107
Key Informatics Challenges in Metabolomics

• Spectra -gt Lists
• Data integrity and quality
• Data alignment and normalization
• Data reduction and classification
• Assessment of significance
• Metabolite identification/quantification
• Lists -gt Pathways Biomarkers
• Pathway mapping and identification
• Biological interpretation