GARRET CUME Flashcards
Define Mass resolution
For two peaks of equal height M2 -m1 is about full width half max of either peak alone
What is mass resolving power
m2 / (m2-m1) where m2-m1 is the mass resolution
(so can also be m / (FWHM) for 2 peaks equal height
What are some excpetions for resolution and resolving power
Multiply charged ion - m is replaced by m/z
What is considered “HIGH MASS RESOLVING POWER”
greater than 10k
What is RMS difference
error in ppm
Mass measurement precision - what is it
m/rms(error in ppm - from a large number of repeated measurements (so can be increased by sampling more data
Talking about mass measurement precision why are TOF’s so accurate?
Do to their fast digitizers means more points per peak width which means more sampling - = better mass measurement precision
External vs internal calibration
2x3 times less accurate than internal calibration
What resolving power do you need to resolve things of the same nominal mass but different elemental comp
200 k
What are some major factors that make TOF usable today as a high res high accuracy intstrument
Delayed ion extraction to deal with KE spread
Fast digitizers
MALDI and MALDI imaging
Orthogonal acceleartion to deal with positional and velocity spread (also ion accumulation)
Reflectron/Multipass - longer flight time means more resolution T/2dt
What is multipass TOF
has 4 electric sectors and essentially the ion goes in an elliptical shape and can do a number of turns (eg 500) can achieve 130k resolution but has limited mass range
What are the major advancements in TOF detectors
going from MCP to things that have more sensitivity at higher mass range (MCP are based on velocity and higher mass have lower velocity) - so these nanomembrane, nanomechanicl resonator and cryogenic detectors are better at this high mass
also TOF TOF so high resolution first analyzer
Where does TOF shine in comparison to other MS
No theoretical upper limit for M/Z
also fast response/scan rate
Applications for TOF?
Tox screening - fast scan speed good for LC-ESI TOF MS, additionally most tox applications are low res so this is an improvement
MSI - again fast scan speed good here - good sensitivity
Aerosol analysis - again coupled to 2d GC - really SMALL /fast peaks so fast rate is key here
Polymers - top down analysis - HIGH mass range key here - can USE ESI for intact top down approach, can do MS^2 in QTOF,
Basics on how FT and orbitrap work
1) Produce ION packets
2)CYCLOTRON motion from the magnet traps radial - brought into coherence with dipolar RF excitation
3)orbitrap achieved by injecting in a short time compared to ion oscillation frequency
orbi is 2 electrodes one around the other - in ideal kingdom field the axial motion is m/z dependant -
- ICR spatial coherence maintained through confining with magnetic and electrostatic fields whereas orbitrap does it and radial coherence is lost as it form a rotating ring
4) detection done from the periodic motion - from detector plates creates an IMAGE CURRENT
5) detect ions - time domain subject to transformation - frequency domain then to mass
6) resolution based on transient time
Major advances in FT’s and orbis
FT - using higher field strenght magnet (15 and 21 T)
Orbi - based off equation can either change ratio of R2 and R1 (radius of outer barrel vs ratio of inner spindle)
Both also have geometries that can be optimized/overcome imperfections from requires ions to pass through
better calibration equations to deal with ion abundance/systematic error (Adding an ion abundance term
Phase correction (using absorption vs magnitude)
Imrpoved data stations with lower latency ,
Other FT improvements
Phase correction(?) dont understand need to read again
Also Data workstations have gotten lower latency - allow for more control to increase res
Ways Fragmentation can be done in FT and orbi
CID or ETD for orbi, outside
CID (internal or external) , ECD, IRMPD in FT
FT and orbi applications
Top down proteomics - High res and mass accuracy really key to resolve isotope distributions from peptides, high charge states and isobaric peptide compositions (can also use LC-ESI for more fractionation or MALDI for high mass, less charged ions)
also good for bottom up and shotgun
middle down protoemics (for large proteins that do not fragment in gas phase efficiently - do partial protein digestion - LARGE peptides
Protein quant - SILAC,
Metabalomics
Lipidomics - elemental composition from isotope fine structure
RNA/DNA analysis - can look at synthetic, RNA sequencing, can also see interferences s, tRNA PTMS, etc - looking at nucleic acid contaminants of asphaltene during crude oil processing
ENviro and food safety - trace contaminants, analogs etc c
Difference between bottom up and shotgun
Bottom is seperated on a gel first, shotgun isnt
Strengths and challenges of metabolomics vs proteomics
Strengths -
many metabolites can be common across species
Dynamics fast - reflect changes well
Sample prep relatively simple (can be more complex if needed)
Challenges:
False discovery rates difficult to ascertain, ID cannot be inferred from fragments, many metabolites being common can make it discern biological origin in a study
Targeted vs untargeted appraoch/goals
Targeted is typically quantification/detection of a known limited panel (hypothesis driven)- validation , absolute quant, ms/ms compared to standard
Untargeted is not limited- as many as possible - acquire features - ID them and try to review known and unknown metabolic changes (hypothesis generating)- discoveryrelative ,
biased by sample prep.
quant ms/ms compared to library qualitative ID
2 approaches to untargeted metabolomics
DDA - data dependant we scan Q1 and then go back and do MS/MS of the highest intensity ions
- cleaner spectra - bias for high intesnity
DIA - Data independent - scan Q1 and then MS/MS of everything
various strategies such as MS^E - do everything or SWATH - do in m/z buckets
a lot messier but gets everything somewhat intensity independent
Common instrumentation for metabolomics
LC, I/M, GC, CE, MS or NMR
LC–MS is popular , polar non polar, sensitivity, lots of good literature,
I/M becoming more popular for enantiomers etc
What are some strengths of ion mobility
increased peak capacity
faster than chromatography
determine CCS - another way to fractionate or separate -
hsycial property of molecule not setting dependant
More precise - inter lab CCS is less than 5% for awide range of molecules
General steps in Metabolomics data anlysis/ID
General steps
ncluding noise filtering, peak detection, peak deconvolution, retention time alignment, and finally
feature annotation.
What are the 5 levels of confidence
LVL 5- UNIQUE feature (eg mass measurement)
LVL 4 - Molecular FOrmula (charge state, isotope abundance , high accuracy)
LEVEL 3 - Tentative structure (MS1 m/z database match) -accurate mass and isotopic structure produce data base matches
LEVEL 2 - putative ID - MS/MS spectrum match (structura info))(almost like level1 without a reference standard - orthogonal info -mainly MS/MS)
LEVEL 1 - validated ID - reference standard confirmation with 2 orthogonal pieces of info
How to calculate mass defect
Exact monoisotopic mass - nominal mass (just adding up protons and neutrons)
How to do you make a kendrick plot
Multiply the m/z’s by (CH2 nominal)/(CH2exact) and that is the kendrick mass. We then calculate Kendrick mass defect which is Kendrick exact - Kendrick nominal (essentially just getting mass defect of the non CH2 pieces) and then plot Kendrick mass defect on y axis vs Kendrick nominal mass
What is the purpose of kendrick mass analysis
To greatly simplify a complex spectra of many different types of the same compound (eg hydrocarbons) - can show on a 2d plot - instead of showing each molecule as is on a mass spectra - reduce it to just the mass and mass defect of differences - so can easily group different hydrocarbons and display on plot (those of the same type will all be on the same row with the same mass defect
so can sort by things like alkylation, heteroatom, (N,O,S) and double bond equiavalnets
How can kendrick analysis be expanded
Instead of multiplying to turn alll of our CH2 nominal - can do with oxygen if differ by oxidation, or halogenation (multiply nominal/exact) (chlorination, bromination)
WHAT is mass defect filtering and uses
exatly what it sounds like -we can filter by mass defect and hopefully just get the class of compounds in our spectra - goo[d for very complex samples like crude oil extracts - removing noise and background peaks, especially when a lot of the components might have similar polarity and are difficult to separate by chromatography
How fo you calculate relative mass defect
RMD, expressed
in ppm, is calculated by dividing the mass defect by the monoi-
sotopic mass multiplied by 10 6
- It’s use is in showing elemental composition eg % hydrogen (has a positive mass defect of 7 ish) -s o can be another way to organize data - lipdis have a range of RMD, alkanes have a nage, sugars have their own range - etc
Whats a major factor in amtrix for drug metabolism DMF
complexity/interferences - they found Urine didnt work well - MDF alone not sufficient to remove all interferents in urine
Primary vs other metabolites
primary are made by the body - synthesis encoded by host genome and essential
the others are also essential but not encoding in genome -(eg get exogenously)
2ndary metabolites are ones that come exogenously but ARE NOT ESSENTIAL eg pollutants, herbicide, pesticide etc
endogenous vs exogenous metabolite cmmonality?
endogeoous are super common across most species, exogenous vary greatly
What is the metabotype
Essentially the metabolic readout of phenotype
What are the 4 metabolic experiments and describe
1) Targeted metabolomics -ID and quant- hypothesis testing HIGH throughput
2) Untargeted – hypothesis generating , generate/collect features, discovery (LC-MS and NMR for ID)
3) Fluxomics – measuring how metabolites change – measure eraectionr ates etc – movement of isotopic labels through intermediates (dynamic!)
4) Metabolite imaging -NMR, MRS, PET,, MALDI, DESI , Imaging MS
What are examples of metabolomic features?
isotope, salt adducts, RT, CCS, dimers, multiply charged, fragmentation pattern, accurate mass, isotope fine structure,
What are the poitns of PCA and PLS-DA
clustering factors to know the key factors that are changing over time
Footprint vs fingerprinting
fingerprinting is looking at cell cultures or blood
footprinting - looking at biological excretions like urine or sweat
Pros and cons of different metabolic samples
cells non invasive but immortal isolated -
organs are great very in tune with specific parts of physiology and body systems but need to be dead
biofluids are pretty good proxies for organs and tissues can be localized ish
what are IEM
inborn errors of metabolism - congenital disorder of metabolism - use of metabolomics to diagnose - have 13 categories like carbs, amino acid, urea cycle, organic acid metabolism etc
What are metabotoxins?
Endogenous molecules like hormones steroids etc, that cause abnormal IEMS etc (endocrinology based) (level based)
10 hallmarks of cancer
According to Hanahan and
Weinberg (188), there are 10 hallmarks of cancer, includ-
ing 1) dysregulated cellular energetics, 2) resistance to
cell death, 3) genome instability, 4) induction of angio-
genesis, 5) metastasis and cell invasion, 6) tumor-pro-
moting inflammation, 7) replicative immortality, 8)
avoiding immune destruction, 9) avoiding growth sup-
pressors, and 10) sustained proliferative signaling
regroup as things that are dysregulated:
Energetics, genome,
Things that are encouraged:
Inflammation, metastasis/cell invasion, proliferative signalling, angiogenesis
THings that are avoided/resisted:
Cell death, immune destruction, growth supressors
and
REplicative immortality
Talk about mass defect labeling
We can attach cunfctional groups with chlorine, bromine, iodine and arsenic as these can move the mass defect into a unique region where there are few species allowing us to better see our analyte (most proteins are C, N’s and H’s so having something unique really pushes it out there to less snosiy region)
Can be really useful in messy top down experiments
or for picking out peaks on labeled regions eg N or C termini (and by consequence and fragment containing those) can also label Cysteines for in sequence labeling
Can also just act as labels - adding functionality- changing MS/MS etc
Hydrogen effect on RMD
RMD is a good reflects the fractional hydrogen content due to hydrogen having such a large positive mass defect 7.825 mDa
Common RMD ranges
alkanes are 1000, lipids and steroid 600-1000, sugars 300-400, organic acids less than 300
What are the types of things you can see in kendick plot groupings
different hetero atoms, methylation, hdyration, oxidation etc
An example of Kendrick mass for chlorine
halogenated dioxins and gfurans - have a lot of repeating chlorine units so can multiply it by a factor of the chlorine nominal / chlorine isotopic - could group them by the same molecule but with more or less substitued chlorine and if substitued for bromine or different shape higher on chart
Uses for mass defect filtering
1) Can filter out things that might have a similar mass but a very different mass defect based on its composition - so good for very messy/convoluted data (depending on hydrogen content can have unique mass defect or phosphopeptides with uniquely negative mass defect
2) Another example is proteins if they make unique ions (like immonium ions) or have phosphates or make unique fragemtns can make them easier to pick out
3) Another example is instead of just grouping naturally occurring molecules can group compounds in your sample like proteins - grouped by mass defect can indicate things about their structure(eg basic, aliphatic, acidic)
4) Drugs - if they have common adducts/ metabolism steps can make dynamic filters that combine mass AND mass defect (drug + adduct for mass and mass defect)bas
What is the current state of false discovery rate in metabolomics and how is it determined
currently no standard - one method is to create a decoy library and see matches made by the methodology
What types of statistical tests/approaches are used at varying levels of confidence in metabolomics
Level 5 and 4 (m/z and molecular formula - can use t test, PCA, PLS, -group data - at this stage we’re ranking significant differences/prioritizing data
Level 3 - - at this stage matching a parent with a specific frag pattern (so the approach/treatment is MS1 and MS/MS data base/ibrary matching)
Level 2 and 1 - integration with known biology is the goal - this uses PATHWAY analysis
4 steps of metabolomic analysis
Biological sample - Metabolically quench, - extraction- chem analysis - Data analysis
4 types of metabolomics experiments
1) Targeted metabolomics -ID and quant- hypothesis testing HIGH throughput
2) Untargeted – hypothesis generating , generate/collect features, discovery (LC-MS and NMR for ID)
3) Fluxomics – measuring how metabolites change – measure eraectionr ates etc – movement of isotopic labels through intermediates (dynamic!)
4) Metabolite imaging -NMR, MRS, PET,, MALDI, DESI , Imaging MS
Some common types of metabotoxins
diabetogens, artherotoxins, obesogens etcnephrotoxin
Talk about how metabolomics can look at disease specifically with chronic kidney disease
First off - Chronic kidney disease has it’s own factors that lead to it BUT then it’;s dysfunction causes a whole new set of metabolties/molecules that can cause a whole host of other diseases (such as atherosclerosis, hypertension, stroke diabetes , cancer etc.
They call the compounds that cause these diseases as uremic toxins - slow acting metabotoxins generated from renal failure and build up over time
Note these can be common in the diet but it’s specifically their level/concentration which is harmful
As such - these uremic toxins play a key role in chronic kidney disease but can accumulate in other ways and be part of those pathways too
Can then additionally look at the Genome and SNPs of these molecule if certain genes directly relate to their basal rate and whether there are genetic factors that predispose someone for higher or lower concentrations of these uremic toxins
Talk about timing for metabolomics of disease
longitudinal studies are a lot better for applicability as can see time course of pre , during post etc and how metabolites change through out - these are different apnels of metabolites as well(pre cursors, ones that are causing the disease etc)
ideally cross sectional and longitudinal - metabalomics is one piece of the puzzle fitting into other types of data
Whats a secretagogue and how might they play a role in diabetes
A substancte that causes another to be secreted - so these sdecretagogue metabolites can cause insulin or glucagon secretions causing the body to mimc chronically high insulin -which trains /builds an inborn insensitivity to normal insulin levels which can be a factor leading to Type 2 diabetes
Talk about metabolomics in cancer what are some major factors to consider?
Onco metabolites - can see the scheme of discovery and then finding out they play roles in cellular metabolism and particularly they change metabolism such that it promotes tumor growth
The warburg effect - is the idea that tumors consume more glucose than normal cells -i.e. they have their own specific changed metabolism
SO we can look at these oncometabolties that cause these changes
OR
in fact the metabolites caused by this new tumor promoting metabolism (cataolic building up)
ANd this can kind of be applied to all 10 hallmarks of cancer (see other question)
6 major classes of small molecule hormones
amino acids, biogenic amins, eiconasid , organic acids, steroids and sugar
Talk about the basis for gut metabolomics
gut metabolome well defined - large intestine leads to water absorption but also has microbes/bacertia 500-1000 and MOST IMRPOTANTLY - they are acting on and processing chemicals which means they have their own metabolome which can be very unique depending on diet etc- incredibly diverse
What sample can we look at for gut metabolome
Feces - but also urine - sicne the water form the large intestine gets reabsorbed - these molecules/baceteria can be reabsorbed and expelled through urine
What are the various Axes and what do they mean
Gut brain axis, - large source of neurotransmitters from the gut
Gut liver axis- largely bile acids (formed from bacteria actin on bile salts) that modulate TG metabolism, glucose metabolism and liver growth
Gut lung axis- Lack of early childhood explore to infectious agents or gut microbe increases causes increased susceptibility to allergic diseases (eg clostridium is one seen to really help)
Gut-Kidney - can see epinephrine and norepinephrine play a role in bacteria growth/attachment in the large intestine
Essentially the interplay between these two systems - how do the metabolomes interact and effect each other
Genomes and SNPs in emtabolomics
Good for integrating metbaolomics data
Also good for seeing what natural abasal rate for some of these molecules because often these molecules themselves are not dangerous or prescurosrs to disease but their relative concentration (a lot of these can be common or in common pathways but if upregulated/downregulated for whatever reason)
How is chlorine subsitution used in kendrick plots
instead of just doing chlorine nominal/chlorine monisotope THey used chlorine substituting hydrogen (so additional of a chlorine - hydrogen) SO the factor you multiply by is (chlorine - hydrogen - nominal)/chlorine - hydrogen (monoisotopic)
go through more protein related things formass defect analysis
stable immonion generated with large mass defect can be identified (from phosphotyrosine) - , modification of phosphor amino acids - unique ions an unique mass defects
CYPAA -
What is mummichog
use Ms 1 and sig measurements to predict metabolic activity networks with MS MS based on changes etc - various platforms doing this predictive or based on data from multiple sample s
7 golden rule
restrict # of elements
follow lewis and senior rules
Isotope patterns
C/H ratio
Heteroatoms to carbon ratio
element ratio probability
TMS consideration
Basicall it should be reasonable - in all these regards
If we have a decooy library that violates this shouldn’t identify any of these as legit
What are apodization functions
functions that help ID small peaks next to a large peak by thinning them out
