Model/cell line selection depends on several factors:

Biological Relevance:

• Does the model accurately represent the disease/condition being studied?
• Does it express the proteins of interest?
• Is it from the relevant tissue type?

Technical Considerations:

• For SILAC: Cells must be able to grow in culture and incorporate labeled amino acids
• Protein yield: Sufficient protein for analysis
• Reproducibility: Well-characterized, stable cell lines preferred
• Availability: Commercially available vs. primary cells

Common choices:

• HeLa cells: Easy to culture, well-characterized
• HEK293: High transfection efficiency
• Primary cells: More physiologically relevant but harder to work with
• Patient-derived cells: Most relevant for translational studies

Pooled Samples:

• Multiple individual samples combined into one
• Represents an "average" of the group
• Advantages:
  • Reduces individual biological variation
  • Increases protein amount for analysis
  • Reduces number of MS runs needed
  • Cost-effective for initial screening
• Disadvantages:
  • Loses individual variation information
  • Cannot identify outliers
  • Cannot perform statistical analysis on individuals

Single/Individual Samples:

• Each sample analyzed separately
• Advantages:
  • Captures biological variability
  • Enables proper statistical analysis
  • Can identify individual responders/non-responders
  • Required for biomarker validation
• Disadvantages:
  • More expensive (more MS runs)
  • More time-consuming
  • May have limited sample amount per individual

ESI Mechanism (step-by-step):

1. Spray Formation: Sample solution is pumped through a capillary needle at high voltage (2-5 kV)
2. Taylor Cone: Electric field causes liquid to form a cone shape at the needle tip
3. Droplet Formation: Fine charged droplets are sprayed from the cone tip
4. Desolvation: Warm nitrogen gas assists solvent evaporation; droplets shrink
5. Coulombic Explosion: As droplets shrink, charge density increases until Rayleigh limit is reached → droplets explode into smaller droplets
6. Ion Release: Process repeats until fully desolvated, multiply charged ions are released

Types of ions formed:

• MULTIPLY CHARGED ions — this is the key characteristic!
• Positive mode: [M+nH]ⁿ⁺ (e.g., [M+2H]²⁺, [M+3H]³⁺)
• Negative mode: [M-nH]ⁿ⁻
• Creates a charge envelope (Gaussian distribution of charge states)

Why multiple charges matter:

• m/z = mass / charge
• Multiple charges reduce m/z values
• Allows large proteins (>100 kDa) to be analyzed within typical mass analyzer range

Advantages of ESI:

• Soft ionization (minimal fragmentation)
• Directly compatible with LC (on-line coupling)
• Very high sensitivity (attomole range)

Disadvantages:

• Sensitive to salts and detergents (ion suppression)
• Requires clean samples
• More complex spectra due to multiple charge states

How MALDI works:

1. Sample Preparation: Analyte mixed with organic matrix (e.g., α-CHCA, DHB, sinapinic acid)
2. Crystallization: Mixture spotted on metal plate; solvent evaporates forming co-crystals
3. Laser Irradiation: UV laser (337 nm nitrogen or 355 nm Nd:YAG) hits the crystals
4. Matrix Absorption: Matrix absorbs photon energy, becomes electronically excited
5. Desorption: Matrix undergoes "micro-explosion," ejecting analyte into gas phase
6. Ionization: Proton transfer from matrix to analyte creates ions

Types of ions:

• SINGLY CHARGED ions — key difference from ESI!
• Positive mode: [M+H]⁺ (most common for peptides)
• Negative mode: [M-H]⁻
• Also: [M+Na]⁺, [M+K]⁺ (adducts)

Pros:

• Simple spectra (singly charged = easy interpretation)
• More tolerant to salts and contaminants than ESI
• Very robust, high-throughput (~10⁴ samples/day)
• Wide mass range (up to 500 kDa)
• Easy to use

Cons:

• Lower sensitivity than ESI (femtomole vs. attomole)
• Not easily coupled to LC (off-line)
• Matrix interference in low mass region
• Shot-to-shot variability

Typical analyzers used with MALDI:

• TOF (Time-of-Flight) — most common combination (MALDI-TOF)
• TOF/TOF — for MS/MS analysis
• Can also be coupled with: FT-ICR, Orbitrap

SELDI (Surface-Enhanced Laser Desorption/Ionization):

A variation of MALDI where the target surface is chemically modified to selectively bind certain proteins.

Key difference from MALDI:

Feature	MALDI	SELDI
Surface	Inert metal plate	Chemically modified (active) surface
Sample prep	Simple spotting	Surface captures specific proteins
Selectivity	None (all proteins)	Surface-dependent selectivity
Complexity	Full sample complexity	Reduced (only bound proteins)
Washing	Not typical	Unbound proteins washed away

SELDI Surface Types:

• Chemical surfaces:
  • CM10: Weak cation exchange
  • Q10: Strong anion exchange
  • H50: Hydrophobic/reverse phase
  • IMAC30: Metal affinity (binds His, phosphoproteins)
• Biological surfaces:
  • Antibody-coated
  • Receptor-coated
  • DNA/RNA-coated

SELDI Workflow:

1. Spot sample on modified surface
2. Specific proteins bind based on surface chemistry
3. Wash away unbound proteins
4. Apply matrix
5. Analyze by laser desorption (same as MALDI)

SELDI Advantages:

• Reduces sample complexity (acts as "on-chip purification")
• Good for biomarker discovery/profiling
• Requires minimal sample preparation

SELDI Limitations:

• Lower resolution than standard MALDI
• Limited protein identification (profiling only)
• Reproducibility issues
• Largely replaced by LC-MS approaches

PMF (Peptide Mass Fingerprinting):

A protein identification method where a protein is enzymatically digested into peptides, and the resulting peptide masses are compared to theoretical masses from database proteins.

PMF Workflow:

1. Protein isolation: Usually from 2D gel spot
2. Destaining: Remove Coomassie/silver stain
3. Reduction & Alkylation: Break and block disulfide bonds
4. Enzymatic digestion: Typically with trypsin
5. Peptide extraction: From gel pieces
6. MALDI-TOF analysis: Measure peptide masses
7. Database search: Compare experimental masses to theoretical

Digestion enzyme — TRYPSIN:

Specificity:

• Cleaves at the C-terminal side of:
• Lysine (K) and Arginine (R)
• EXCEPT when followed by Proline (P)

Why trypsin is the gold standard:

• High specificity: Predictable cleavage sites
• Optimal peptide size: 6-20 amino acids (ideal for MS)
• Basic residues at C-terminus: Promotes ionization in positive mode
• Robust: Works well across pH 7-9
• Reproducible: Produces consistent results
• Self-digestion peaks: Can be used for internal calibration

Other enzymes sometimes used:

• Chymotrypsin: Cleaves after Phe, Tyr, Trp
• Glu-C: Cleaves after Glu (and Asp at high pH)
• Lys-C: Cleaves after Lys only
• Asp-N: Cleaves before Asp

Important clarification: Shotgun proteomics IS a type of bottom-up approach. The distinction is in the workflow:

Feature	Classical Bottom-Up (PMF)	Shotgun (Bottom-Up)
Protein separation	FIRST (2D-PAGE, then cut spots)	None or minimal
Digestion	Single isolated protein	Entire protein mixture
Peptide separation	Usually none	LC (often multi-dimensional)
MS analysis	MALDI-TOF (PMF)	LC-MS/MS
Identification	Mass matching	MS/MS sequencing
Throughput	One protein at a time	Thousands of proteins

Classical Bottom-Up Workflow:

1. Separate proteins by 2D-PAGE
2. Cut out individual spots
3. Digest each spot separately
4. Analyze by MALDI-TOF
5. PMF database search

Shotgun Workflow:

1. Lyse cells, extract all proteins
2. Digest entire mixture into peptides
3. Separate peptides by LC (MudPIT uses 2D-LC)
4. Analyze by MS/MS
5. Database search with MS/MS spectra

Why "Shotgun"?

• Like a shotgun blast — analyzes everything at once
• No pre-selection of proteins
• Relies on computational deconvolution

Sample-Related Limitations:

• Hydrophobic proteins: Membrane proteins poorly soluble in IEF buffers → underrepresented
• Extreme pI proteins: Very acidic (<3) or basic (>10) proteins difficult to focus
• Extreme MW proteins:
  • Large proteins (>200 kDa) don't enter gel well
  • Small proteins (<10-15 kDa) may run off the gel
• Low-abundance proteins: Masked by high-abundance proteins; below detection limit
• Dynamic range: Limited (~10⁴), much less than proteome range (~10⁶-10⁷)

Technical Limitations:

• Poor reproducibility: Gel-to-gel variation requires running in triplicate
• Labor-intensive: Manual, time-consuming, hard to automate
• Low throughput: Cannot be easily scaled up
• Co-migration: Proteins with similar pI/MW appear in same spot
• Quantification limited: Staining is semi-quantitative at best

Analytical Limitations:

• Proteome coverage gap: Yeast example: 6,000 genes → 4,000 expressed proteins → only ~1,000 detected by 2DE
• Requires MS for ID: 2DE is only separation; identification needs additional steps
• PTM detection: May see multiple spots but hard to characterize modifications

Practical Issues:

• Streaking/smearing from degradation
• Background interference from staining
• Keratin contamination common

Hybrid MS: Instruments combining two or more different mass analyzers to leverage their complementary strengths.

Common Hybrid Configurations:

Hybrid Type	Components	Strengths
Q-TOF	Quadrupole + TOF	High resolution, accurate mass, good for ID
Triple Quad (QqQ)	Q1 + Collision cell + Q3	Excellent for quantification (SRM/MRM)
Q-Orbitrap	Quadrupole + Orbitrap	Very high resolution + sensitivity
LTQ-Orbitrap	Linear ion trap + Orbitrap	High speed + high resolution
TOF-TOF	TOF + Collision + TOF	High-energy fragmentation with MALDI
Q-Trap	Quadrupole + Ion trap	Versatile, MRM + scanning modes

How they work (Q-TOF example):

1. Q1 (Quadrupole): Selects precursor ion of interest
2. Collision cell: Fragments the selected ion (CID)
3. TOF: Analyzes all fragments with high resolution and mass accuracy

Limitations of Hybrid Systems:

• Cost: Very expensive instruments ($500K - $1M+)
• Complexity: Requires expert operators
• Maintenance: More components = more potential failures
• Data complexity: Generates massive datasets
• Duty cycle trade-offs: Can't optimize all parameters simultaneously
• Ion transmission losses: Each analyzer stage loses some ions

Specific limitations by type:

• Q-TOF: Lower sensitivity in MS/MS mode
• Ion trap hybrids: Space charge effects limit dynamic range
• Orbitrap hybrids: Slower scan speed than TOF

TUNEL = Terminal deoxynucleotidyl transferase dUTP Nick End Labeling

Purpose: Detects apoptosis (programmed cell death) by identifying DNA fragmentation.

Principle:

• During apoptosis, endonucleases cleave DNA between nucleosomes
• This creates many DNA fragments with exposed 3'-OH ends ("nicks")
• TUNEL labels these free 3'-OH ends

How it works:

1. TdT enzyme (terminal deoxynucleotidyl transferase) is added
2. TdT adds labeled dUTP nucleotides to 3'-OH ends of DNA breaks
3. Labels can be: fluorescent (FITC), biotin (detected with streptavidin), or other markers
4. Visualized by fluorescence microscopy or flow cytometry

Applications:

• Detecting apoptosis in tissue sections
• Studying cell death in disease models
• Drug toxicity testing
• Cancer research

Limitations:

• Can also label necrotic cells (not specific to apoptosis)
• False positives from mechanical DNA damage during sample prep
• Should be combined with other apoptosis markers

Follow-up study suggestions:

• Caspase activity assays (more specific for apoptosis)
• Annexin V staining (early apoptosis marker)
• Western blot for cleaved caspase-3 or PARP

Phage Display: A molecular biology technique where peptides or proteins are expressed ("displayed") on the surface of bacteriophage particles.

How it works:

1. Library Creation: DNA encoding peptides/proteins is inserted into phage coat protein gene
2. Expression: Phage expresses the foreign peptide fused to its coat protein (usually pIII or pVIII)
3. Panning: Library exposed to target molecule (bait) immobilized on surface
4. Selection: Non-binding phages washed away; binding phages retained
5. Amplification: Bound phages eluted and amplified in bacteria
6. Iteration: Process repeated 3-4 times to enrich for strong binders
7. Identification: DNA sequencing reveals the binding peptide sequence

Applications:

• Antibody discovery and engineering
• Finding protein-protein interaction partners
• Epitope mapping
• Drug target identification
• Peptide ligand discovery

MAIN LIMITATIONS:

• Bacterial expression system:
  • No post-translational modifications (no glycosylation, phosphorylation)
  • May not fold mammalian proteins correctly
  • Codon bias issues
• Size constraints: Large proteins difficult to display
• Selection bias: Some peptides toxic to bacteria → lost from library
• False positives: Selection for phage propagation, not just binding
• Context-dependent: Displayed peptide may behave differently than free peptide
• Limited to protein/peptide interactions: Cannot study interactions requiring membrane context

Energy Transfer Methods: Techniques that detect protein-protein interactions based on the transfer of energy between two labeled molecules when they come into close proximity.

FRET (Förster Resonance Energy Transfer):

• Donor: Fluorescent molecule that absorbs excitation light (e.g., CFP, GFP)
• Acceptor: Fluorescent molecule that receives energy from donor (e.g., YFP, RFP)
• Mechanism: Non-radiative energy transfer through dipole-dipole coupling
• Distance requirement: 1-10 nm (typically <10 nm for efficient transfer)

BRET (Bioluminescence Resonance Energy Transfer):

• Donor: Bioluminescent enzyme (e.g., Renilla luciferase)
• Acceptor: Fluorescent protein (e.g., GFP, YFP)
• Advantage: No external excitation needed → lower background

Signals Obtained:

• When proteins are FAR apart:
  • Only donor emission observed
  • No energy transfer
• When proteins INTERACT (close proximity):
  • Donor emission decreases (quenching)
  • Acceptor emission increases (sensitized emission)
  • FRET efficiency can be calculated

Types of signals measured:

1. Sensitized emission: Acceptor fluorescence upon donor excitation
2. Donor quenching: Decrease in donor fluorescence intensity
3. Donor lifetime: Decrease in fluorescence lifetime (FLIM-FRET)
4. Acceptor photobleaching: Donor recovery after acceptor is bleached

Applications:

• Detecting protein-protein interactions in living cells
• Monitoring conformational changes
• Studying signaling pathway activation
• Biosensor development

Common FRET pairs:

• CFP (cyan) → YFP (yellow)
• BFP (blue) → GFP (green)
• GFP → RFP/mCherry

CID (Collision-Induced Dissociation): A fragmentation method where precursor ions are fragmented by colliding them with an inert gas.

How CID works:

1. Ion selection: Precursor ion selected in first mass analyzer (MS1)
2. Collision cell: Selected ion enters a chamber filled with inert gas (Argon, Nitrogen, or Xenon)
3. Collision: Ion collides with gas molecules, converting kinetic energy to internal energy
4. Fragmentation: Internal energy causes bonds to break, producing fragment ions
5. Analysis: Fragment ions analyzed in second mass analyzer (MS2)

Significance in MS/MS:

• Generates b-ions and y-ions for peptide sequencing
• Provides structural information about the parent ion
• Enables amino acid sequence determination
• Allows protein identification via database searching
• Can reveal PTM locations

Other fragmentation methods:

• HCD: Higher-energy Collisional Dissociation (used in Orbitrap)
• ETD: Electron Transfer Dissociation (better for PTMs, larger peptides)
• ECD: Electron Capture Dissociation (preserves labile modifications)

Fragment ions from peptide backbone cleavage:

b-ions:

• Contain the N-terminal portion of the peptide
• Charge retained on the N-terminal fragment
• Named b₁, b₂, b₃... (number = amino acids from N-terminus)

y-ions:

• Contain the C-terminal portion of the peptide
• Charge retained on the C-terminal fragment
• Named y₁, y₂, y₃... (number = amino acids from C-terminus)

Visual representation:

        N-terminus ← → C-terminus
        H₂N-[AA₁]-[AA₂]-[AA₃]-[AA₄]-COOH
             ↓     ↓     ↓
            b₁    b₂    b₃    (N-terminal fragments)
                  y₃    y₂    y₁  (C-terminal fragments)
        

How sequencing works:

1. Mass differences between consecutive b-ions (or y-ions) = amino acid masses
2. b₂ - b₁ = mass of 2nd amino acid
3. y₃ - y₂ = mass of amino acid at position (n-2)
4. Complete series allows full sequence determination

Why both series are useful:

• Complementary information confirms sequence
• Gaps in one series may be filled by the other
• b + y should equal precursor mass + 18 (water)

Monoisotopic Mass:

• Mass calculated using the most abundant isotope of each element
• For organic molecules: ¹²C, ¹H, ¹⁴N, ¹⁶O, ³²S
• Corresponds to the first peak in the isotope distribution (M+0)
• More precise, used for accurate mass measurements

Average Mass:

• Weighted average of all naturally occurring isotopes
• Takes into account natural isotope abundance
• Corresponds to the centroid of the isotope envelope
• Used when resolution is insufficient to resolve isotopes

Example (for Carbon):

• Monoisotopic: ¹²C = 12.0000 Da
• Average: (98.9% × 12.0000) + (1.1% × 13.0034) = 12.011 Da

Use in PMF:

Situation	Mass Type	Reason
High-resolution MS (MALDI-TOF)	Monoisotopic	Can resolve isotope peaks
Low-resolution MS	Average	Cannot resolve isotopes
Small peptides (<2000 Da)	Monoisotopic	First peak is tallest
Large proteins (>10 kDa)	Average	Monoisotopic peak too small to detect

How each analyzer separates ions:

TOF (Time-of-Flight):

• Ions accelerated through same voltage, gain same kinetic energy
• KE = ½mv² → lighter ions travel faster
• Measures flight time through drift tube
• Shorter time = lower m/z

Quadrupole:

• Four parallel rods with oscillating RF/DC voltages
• Creates oscillating electric field
• Only ions with specific m/z have stable trajectories
• Others collide with rods and are lost
• Acts as a mass filter (scanning or SIM mode)

Orbitrap:

• Ions trapped orbiting around central spindle electrode
• Oscillate axially with frequency dependent on m/z
• Measures oscillation frequency (image current)
• Fourier transform converts frequency → m/z

Comparison table:

Parameter	TOF	Quadrupole	Orbitrap
Resolution	10,000-60,000	1,000-4,000 (low)	100,000-500,000+
Mass Accuracy	5-20 ppm	100-1000 ppm	<2-5 ppm
Sensitivity	High (femtomole)	High	High (attomole)
Mass Range	Unlimited (in principle)	Up to ~4000 m/z	Up to ~6000 m/z
Scan Speed	Very fast	Fast	Slower
Cost	Moderate	Low	High
Best for	MALDI, fast scanning	Quantification (SRM)	High accuracy ID

De Novo Sequencing: Determining the amino acid sequence of a peptide directly from its MS/MS spectrum, without relying on a sequence database.

How it works:

1. Acquire high-quality MS/MS spectrum
2. Identify b-ion and y-ion series
3. Calculate mass differences between consecutive peaks
4. Match mass differences to amino acid residue masses
5. Build sequence from N- to C-terminus (or reverse)
6. Validate with complementary ion series

When to use de novo sequencing:

• Protein NOT in database:
  • Novel organisms without sequenced genomes
  • Uncharacterized proteins
  • Organisms with incomplete proteome databases
• Unexpected modifications: PTMs not predicted by database
• Mutations/variants: Sequence differs from database entry
• Antibody sequencing: Highly variable regions
• Ancient proteins: Paleoproteomics
• Validation: Confirming database search results

Challenges:

• Requires high-quality spectra with complete ion series
• Isobaric amino acids (Leu/Ile = 113 Da) cannot be distinguished
• Labor-intensive and time-consuming
• May have gaps in sequence coverage

Software tools: PEAKS, Novor, PepNovo, DeNovoGUI

Inteins: Self-splicing protein segments that can excise themselves from a precursor protein, leaving behind the flanking exteins joined together.

Terminology:

• Intein: INternal proTEIN (gets removed)
• Extein: EXternal proTEIN (flanking sequences that remain)
• N-extein — [INTEIN] — C-extein → N-extein—C-extein + free intein

Mechanism (protein splicing):

1. N-S or N-O acyl shift at N-terminus of intein
2. Transesterification
3. Asparagine cyclization releases intein
4. S-N or O-N acyl shift joins exteins with native peptide bond

Applications in protein engineering:

• Self-cleaving affinity tags:
  • Protein fused to intein + affinity tag (e.g., chitin-binding domain)
  • Bind to affinity column
  • Induce intein cleavage (pH, temperature, or thiol)
  • Pure protein released, tag remains on column
  • Advantage: No protease needed, no extra residues left
• Protein ligation (Expressed Protein Ligation):
  • Join two protein fragments with native peptide bond
  • Useful for incorporating unnatural amino acids
  • Creating segmentally labeled proteins for NMR
• Protein cyclization: Create cyclic proteins
• Conditional protein splicing: Control protein activity

Interactomics: The study of protein-protein interactions (PPIs) and the networks they form within biological systems.

1. Yeast Two-Hybrid (Y2H):

• Principle: Reconstitution of transcription factor activity
• Method:
  • Bait protein fused to DNA-binding domain
  • Prey protein fused to activation domain
  • If bait and prey interact → transcription factor reconstituted → reporter gene expressed
• Pros: High-throughput, detects direct binary interactions
• Cons: In vivo but in yeast (not native environment), high false positive rate, only nuclear interactions

2. Co-Immunoprecipitation (Co-IP):

• Principle: Antibody pulldown of protein complexes
• Method:
  • Lyse cells, add antibody against bait protein
  • Antibody-protein complex captured on beads
  • Wash away non-specific proteins
  • Elute and analyze interacting proteins (Western blot or MS)
• Pros: Detects endogenous interactions, physiological conditions
• Cons: Requires good antibody, may miss transient interactions, cannot distinguish direct from indirect interactions

3. Affinity Purification-Mass Spectrometry (AP-MS):

• Principle: Tagged bait protein pulls down interaction partners
• Method:
  • Express tagged bait protein (FLAG, HA, TAP tag)
  • Lyse cells, capture bait + interactors on affinity resin
  • Wash stringently
  • Elute and identify interactors by MS
• Pros: Unbiased identification, can detect entire complexes
• Cons: Tag may affect interactions, overexpression artifacts, false positives from sticky proteins

Method	Throughput	Direct/Indirect	Environment
Y2H	High	Direct only	Yeast nucleus
Co-IP	Low	Both	Native
AP-MS	Medium	Both	Native (with tag)

Strategies when protein is not in database:

1. De Novo Sequencing:

• Determine peptide sequence directly from MS/MS spectrum
• Calculate mass differences between fragment ions
• Match to amino acid masses
• Build sequence without database reference

2. Homology/Sequence Tag Searching:

• Use short sequence tags from de novo to search related organisms
• BLAST search against broader databases (NCBI nr)
• MS-BLAST: Search with imperfect sequences
• May find homologous protein in related species

3. Error-Tolerant Database Searching:

• Allow for mutations, modifications, or sequence variants
• Search with wider mass tolerance
• Consider unexpected PTMs or SNPs

4. EST/Transcriptome Database Search:

• Use expressed sequence tags (EST) databases
• Search against RNA-seq data from same organism
• May contain unannotated protein sequences

5. Spectral Library Searching:

• Compare experimental spectrum to library of acquired spectra
• May match even without sequence information

6. Genomic Six-Frame Translation:

• If genome is available but not annotated
• Translate genome in all 6 reading frames
• Search MS data against translated sequences

Practical workflow:

1. First: Try error-tolerant search or related species database
2. Second: Perform de novo sequencing on best spectra
3. Third: BLAST de novo sequences against NCBI
4. Fourth: If genome available, try 6-frame translation

2D-PAGE = Two-Dimensional Polyacrylamide Gel Electrophoresis

Principle: Separates proteins by TWO independent properties for maximum resolution.

First Dimension: Isoelectric Focusing (IEF)

• Separates proteins by isoelectric point (pI)
• Uses immobilized pH gradient (IPG) strips
• Proteins migrate until net charge = 0
• High voltage (up to 8000 V), long focusing time

Second Dimension: SDS-PAGE

• Separates proteins by molecular weight (MW)
• IPG strip equilibrated with SDS, placed on gel
• SDS denatures proteins and provides uniform charge
• Smaller proteins migrate faster

Complete workflow:

1. Sample preparation: Lysis, solubilization in urea/thiourea/CHAPS
2. Rehydration: Load sample onto IPG strip
3. IEF: Focus proteins by pI (12-24 hours)
4. Equilibration: Reduce (DTT) and alkylate (IAA) proteins in SDS buffer
5. SDS-PAGE: Separate by MW (4-6 hours)
6. Staining: Coomassie, silver, or fluorescent (SYPRO Ruby)
7. Image analysis: Detect spots, compare gels
8. Spot picking: Excise spots of interest
9. MS analysis: In-gel digestion → MALDI-TOF (PMF) or LC-MS/MS

Dimension	Property	Method	Direction
1st	pI (charge)	IEF	Horizontal
2nd	MW (size)	SDS-PAGE	Vertical