"Activity is perfect, so why can't we use it?"
Sena was confused. IC50 2 nM. Ideal numbers.
"Off-target effects," Mikhail explained.
"Off-target?"
Eiji displayed two proteins on screen. "Target and another protein with similar structure."
"This compound binds to both."
Akira read the numbers. "Target: 2 nM. Off-target: 5 nM."
"Almost the same..."
"No selectivity," Mikhail said. "This causes side effects."
Sena asked. "How do we distinguish them? They're similar."
Eiji overlaid the two structures. "Active sites are similar. But differences exist in surrounding areas."
"Here," Eiji pointed. "Sub-pocket shapes differ."
"Target is wide. Off-target is narrow."
Akira presented strategy. "Exploit this difference."
"Add substituent that fits wide pocket. Won't enter narrow pocket."
Lina modeled. "Add phenyl ring."
"Fits in target. In off-target, causes steric clash."
Sena had hope. "This increases selectivity?"
"Computationally," Akira was cautious. "But won't know without measuring."
Mikhail presented another perspective. "Also look at amino acid sequence differences."
"Target is valine. Off-target is isoleucine."
"Similar but subtly different."
Eiji explained details. "Valine is symmetric. Isoleucine is asymmetric."
"Can exploit this asymmetry to distinguish."
Akira proposed. "Branched substituent. Hard to enter isoleucine pocket."
Sena wrote in notebook. "Steric recognition."
"Yes. Lock and key relationship," Mikhail analogized.
Eiji warned. "But increasing selectivity too much can reduce activity."
"What?"
"Sometimes sacrifice strong interactions to gain selectivity."
Akira supplemented. "Trade-off. Balance of activity and selectivity."
"Where should we compromise?"
Mikhail answered. "Depends on clinically required selectivity."
"10-fold? 100-fold? 1000-fold?"
"Determined by side effect severity and dosage."
Eiji showed data. "Selectivity profile of similar target family."
"Kinase inhibitors. Over 500 kinases exist."
"Measure selectivity against all."
Sena was overwhelmed. "500 types..."
"But only some are important," Akira said. "Kinases involved in toxicity."
"hERG, CYP, major metabolic enzymes."
Mikhail organized. "Selectivity panel. Preferentially check important off-targets."
Sena looked at the structure. "So measure selectivity panel with this modification?"
"Yes. Start with most concerning off-targets."
Eiji displayed a list. "These 3. Structurally similar and toxicity-related."
Akira prioritized. "Measure these first. If cleared, proceed next."
"Stepwise evaluation."
Sena asked. "How is selectivity measured?"
Mikhail explained. "Binding assay. Measure IC50 against each protein."
"IC50 ratio of target to off-target is selectivity."
"100-fold selective means IC50 ratio of 100."
Eiji supplemented. "But also confirm at cellular level."
"Environment differs between purified protein and inside cells."
"So also look at selectivity in cell assays."
Sena understood. "Multi-layered evaluation."
"Yes. In vitro, cells, animals. Confirm stepwise."
Akira set realistic goals. "First, 50-fold selectivity for 3 major off-targets."
"If that clears, next step."
Mikhail encouraged. "Selectivity can be improved by design."
"Find structural differences and exploit them."
Eiji said finally. "The key called selectivity. Create a key that fits only the right lock."
"That's the path to drugs without side effects."
Sena resolved. "I'll create modification proposals and measure."
Akira added. "Even if it fails, we learn. Why selectivity didn't emerge."
"That knowledge benefits next design."
Mikhail smiled. "Pursuit of selectivity. An endless journey."
"But we approach step by step."
The four gazed at the structure. The key called selectivity. Sharpening it was today's task. Acting only on the right target. That's the ideal drug form.