[{"data":1,"prerenderedAt":904},["ShallowReactive",2],{"content-query-R69CdEEikl":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"heading":10,"prompt":11,"tags":15,"files":18,"nav":18,"gallery":19,"body":21,"_type":897,"_id":898,"_source":899,"_file":900,"_stem":901,"_extension":902,"sitemap":903},"/tools/hill-equation-fit","tools",false,"","Hill Equation Fit Calculator","Fit Hill equations online from Excel or CSV data. Estimate EC50, slope, cooperativity, and sigmoidal response curves with AI.","Hill Equation Fit",{"prefix":12,"label":13,"placeholder":14},"Fit the Hill equation","Describe the dose-response analysis you want to run","e.g. fit Hill equation to dose-response data, extract EC50 and Hill coefficient, plot sigmoidal curve with 95% CI",[16,17],"statistics","science",true,[20],"/img/tools/hill-equation-fit.png",{"type":22,"children":23,"toc":886},"root",[24,33,103,136,142,228,234,397,402,408,557,563,687,693,747,753,782,788,808,825,842,876],{"type":25,"tag":26,"props":27,"children":29},"element","h2",{"id":28},"what-is-the-hill-equation",[30],{"type":31,"value":32},"text","What Is the Hill Equation?",{"type":25,"tag":34,"props":35,"children":36},"p",{},[37,39,45,47,66,68,73,75,80,82,87,89,94,96,101],{"type":31,"value":38},"The ",{"type":25,"tag":40,"props":41,"children":42},"strong",{},[43],{"type":31,"value":44},"Hill equation",{"type":31,"value":46}," is a sigmoidal mathematical model that describes how a response variable — enzyme activity, receptor occupancy, drug effect, fluorescence signal — saturates as stimulus concentration increases. It takes the form ",{"type":25,"tag":40,"props":48,"children":49},{},[50,52,58,60,64],{"type":31,"value":51},"Response = Emax × ",{"type":25,"tag":53,"props":54,"children":55},"span",{},[56],{"type":31,"value":57},"C",{"type":31,"value":59},"ⁿ / (EC50ⁿ + ",{"type":25,"tag":53,"props":61,"children":62},{},[63],{"type":31,"value":57},{"type":31,"value":65},"ⁿ)",{"type":31,"value":67},", where ",{"type":25,"tag":40,"props":69,"children":70},{},[71],{"type":31,"value":72},"Emax",{"type":31,"value":74}," is the maximum response, ",{"type":25,"tag":40,"props":76,"children":77},{},[78],{"type":31,"value":79},"EC50",{"type":31,"value":81}," (or ",{"type":25,"tag":40,"props":83,"children":84},{},[85],{"type":31,"value":86},"IC50",{"type":31,"value":88}," for inhibition) is the concentration at half-maximal effect, and ",{"type":25,"tag":40,"props":90,"children":91},{},[92],{"type":31,"value":93},"n",{"type":31,"value":95}," is the ",{"type":25,"tag":40,"props":97,"children":98},{},[99],{"type":31,"value":100},"Hill coefficient",{"type":31,"value":102}," that controls the steepness of the sigmoid. When n = 1 the curve reduces to simple Michaelis-Menten hyperbolic saturation; when n > 1 the curve is steeper (cooperative binding or steep dose-response); when n \u003C 1 it is shallower (negative cooperativity or heterogeneous binding sites).",{"type":25,"tag":34,"props":104,"children":105},{},[106,108,113,115,120,122,127,129,134],{"type":31,"value":107},"The Hill coefficient was first derived to describe the cooperative binding of oxygen to hemoglobin (n ≈ 2.8) — where binding of one oxygen molecule increases the affinity for subsequent molecules. Today the Hill equation is the standard model for ",{"type":25,"tag":40,"props":109,"children":110},{},[111],{"type":31,"value":112},"IC50 determination",{"type":31,"value":114}," in drug discovery (what concentration inhibits 50% of activity?), ",{"type":25,"tag":40,"props":116,"children":117},{},[118],{"type":31,"value":119},"EC50 determination",{"type":31,"value":121}," in pharmacology (what concentration produces 50% of the maximum effect?), ",{"type":25,"tag":40,"props":123,"children":124},{},[125],{"type":31,"value":126},"activation curves",{"type":31,"value":128}," in neuroscience (action potential gating), and ",{"type":25,"tag":40,"props":130,"children":131},{},[132],{"type":31,"value":133},"binding curves",{"type":31,"value":135}," in biochemistry (ligand-receptor or ligand-DNA interactions). The parameters are extracted by nonlinear least-squares fitting using the Levenberg-Marquardt algorithm.",{"type":25,"tag":26,"props":137,"children":139},{"id":138},"how-it-works",[140],{"type":31,"value":141},"How It Works",{"type":25,"tag":143,"props":144,"children":145},"ol",{},[146,185,201],{"type":25,"tag":147,"props":148,"children":149},"li",{},[150,155,157,162,164,169,171,176,178,183],{"type":25,"tag":40,"props":151,"children":152},{},[153],{"type":31,"value":154},"Upload your data",{"type":31,"value":156}," — provide a CSV or Excel file with a ",{"type":25,"tag":40,"props":158,"children":159},{},[160],{"type":31,"value":161},"concentration",{"type":31,"value":163}," column (numeric, any units) and a ",{"type":25,"tag":40,"props":165,"children":166},{},[167],{"type":31,"value":168},"response",{"type":31,"value":170}," column (numeric: percent inhibition, fluorescence, activity, occupancy). Optional: a ",{"type":25,"tag":40,"props":172,"children":173},{},[174],{"type":31,"value":175},"group",{"type":31,"value":177}," or ",{"type":25,"tag":40,"props":179,"children":180},{},[181],{"type":31,"value":182},"replicate",{"type":31,"value":184}," column for multiple curves.",{"type":25,"tag":147,"props":186,"children":187},{},[188,193,195],{"type":25,"tag":40,"props":189,"children":190},{},[191],{"type":31,"value":192},"Describe the analysis",{"type":31,"value":194}," — e.g. ",{"type":25,"tag":196,"props":197,"children":198},"em",{},[199],{"type":31,"value":200},"\"fit Hill equation to columns 'concentration_uM' and 'inhibition_pct'; extract IC50, Hill coefficient, and Emax with 95% confidence intervals; plot on log scale\"",{"type":25,"tag":147,"props":202,"children":203},{},[204,209,211,218,220,226],{"type":25,"tag":40,"props":205,"children":206},{},[207],{"type":31,"value":208},"Get full results",{"type":31,"value":210}," — the AI writes Python code using ",{"type":25,"tag":212,"props":213,"children":215},"a",{"href":214},"https://docs.scipy.org/doc/scipy/reference/generated/scipy.optimize.curve_fit.html",[216],{"type":31,"value":217},"scipy.optimize.curve_fit",{"type":31,"value":219}," to fit the Hill equation and ",{"type":25,"tag":212,"props":221,"children":223},{"href":222},"https://plotly.com/python/",[224],{"type":31,"value":225},"Plotly",{"type":31,"value":227}," to render the sigmoidal curve with data points and confidence bands",{"type":25,"tag":26,"props":229,"children":231},{"id":230},"required-data-format",[232],{"type":31,"value":233},"Required Data Format",{"type":25,"tag":235,"props":236,"children":237},"table",{},[238,262],{"type":25,"tag":239,"props":240,"children":241},"thead",{},[242],{"type":25,"tag":243,"props":244,"children":245},"tr",{},[246,252,257],{"type":25,"tag":247,"props":248,"children":249},"th",{},[250],{"type":31,"value":251},"Column",{"type":25,"tag":247,"props":253,"children":254},{},[255],{"type":31,"value":256},"Description",{"type":25,"tag":247,"props":258,"children":259},{},[260],{"type":31,"value":261},"Example",{"type":25,"tag":263,"props":264,"children":265},"tbody",{},[266,317,365],{"type":25,"tag":243,"props":267,"children":268},{},[269,279,284],{"type":25,"tag":270,"props":271,"children":272},"td",{},[273],{"type":25,"tag":274,"props":275,"children":277},"code",{"className":276},[],[278],{"type":31,"value":161},{"type":25,"tag":270,"props":280,"children":281},{},[282],{"type":31,"value":283},"Stimulus or drug concentration (any unit)",{"type":25,"tag":270,"props":285,"children":286},{},[287,293,295,301,302,308,309,315],{"type":25,"tag":274,"props":288,"children":290},{"className":289},[],[291],{"type":31,"value":292},"0.01",{"type":31,"value":294},", ",{"type":25,"tag":274,"props":296,"children":298},{"className":297},[],[299],{"type":31,"value":300},"0.1",{"type":31,"value":294},{"type":25,"tag":274,"props":303,"children":305},{"className":304},[],[306],{"type":31,"value":307},"1.0",{"type":31,"value":294},{"type":25,"tag":274,"props":310,"children":312},{"className":311},[],[313],{"type":31,"value":314},"10",{"type":31,"value":316}," (µM)",{"type":25,"tag":243,"props":318,"children":319},{},[320,328,333],{"type":25,"tag":270,"props":321,"children":322},{},[323],{"type":25,"tag":274,"props":324,"children":326},{"className":325},[],[327],{"type":31,"value":168},{"type":25,"tag":270,"props":329,"children":330},{},[331],{"type":31,"value":332},"Measured effect",{"type":25,"tag":270,"props":334,"children":335},{},[336,342,343,349,350,356,357,363],{"type":25,"tag":274,"props":337,"children":339},{"className":338},[],[340],{"type":31,"value":341},"5",{"type":31,"value":294},{"type":25,"tag":274,"props":344,"children":346},{"className":345},[],[347],{"type":31,"value":348},"28",{"type":31,"value":294},{"type":25,"tag":274,"props":351,"children":353},{"className":352},[],[354],{"type":31,"value":355},"51",{"type":31,"value":294},{"type":25,"tag":274,"props":358,"children":360},{"className":359},[],[361],{"type":31,"value":362},"89",{"type":31,"value":364}," (%)",{"type":25,"tag":243,"props":366,"children":367},{},[368,376,381],{"type":25,"tag":270,"props":369,"children":370},{},[371],{"type":25,"tag":274,"props":372,"children":374},{"className":373},[],[375],{"type":31,"value":175},{"type":25,"tag":270,"props":377,"children":378},{},[379],{"type":31,"value":380},"Optional: compound or condition label",{"type":25,"tag":270,"props":382,"children":383},{},[384,390,391],{"type":25,"tag":274,"props":385,"children":387},{"className":386},[],[388],{"type":31,"value":389},"Drug A",{"type":31,"value":294},{"type":25,"tag":274,"props":392,"children":394},{"className":393},[],[395],{"type":31,"value":396},"Drug B",{"type":25,"tag":34,"props":398,"children":399},{},[400],{"type":31,"value":401},"Any column names work — describe them in your prompt.",{"type":25,"tag":26,"props":403,"children":405},{"id":404},"interpreting-the-results",[406],{"type":31,"value":407},"Interpreting the Results",{"type":25,"tag":235,"props":409,"children":410},{},[411,427],{"type":25,"tag":239,"props":412,"children":413},{},[414],{"type":25,"tag":243,"props":415,"children":416},{},[417,422],{"type":25,"tag":247,"props":418,"children":419},{},[420],{"type":31,"value":421},"Parameter",{"type":25,"tag":247,"props":423,"children":424},{},[425],{"type":31,"value":426},"What it means",{"type":25,"tag":263,"props":428,"children":429},{},[430,446,461,477,493,509,525,541],{"type":25,"tag":243,"props":431,"children":432},{},[433,441],{"type":25,"tag":270,"props":434,"children":435},{},[436],{"type":25,"tag":40,"props":437,"children":438},{},[439],{"type":31,"value":440},"EC50 / IC50",{"type":25,"tag":270,"props":442,"children":443},{},[444],{"type":31,"value":445},"Concentration producing 50% of maximum effect — the primary potency measure",{"type":25,"tag":243,"props":447,"children":448},{},[449,456],{"type":25,"tag":270,"props":450,"children":451},{},[452],{"type":25,"tag":40,"props":453,"children":454},{},[455],{"type":31,"value":72},{"type":25,"tag":270,"props":457,"children":458},{},[459],{"type":31,"value":460},"Maximum response asymptote — 100% for normalized data, actual maximum for raw data",{"type":25,"tag":243,"props":462,"children":463},{},[464,472],{"type":25,"tag":270,"props":465,"children":466},{},[467],{"type":25,"tag":40,"props":468,"children":469},{},[470],{"type":31,"value":471},"Hill coefficient (n)",{"type":25,"tag":270,"props":473,"children":474},{},[475],{"type":31,"value":476},"Sigmoidal steepness: n=1 (simple hyperbolic), n>1 (cooperative/steep), n\u003C1 (shallow)",{"type":25,"tag":243,"props":478,"children":479},{},[480,488],{"type":25,"tag":270,"props":481,"children":482},{},[483],{"type":25,"tag":40,"props":484,"children":485},{},[486],{"type":31,"value":487},"95% CI on EC50",{"type":25,"tag":270,"props":489,"children":490},{},[491],{"type":31,"value":492},"Precision of the potency estimate — narrower CI = better-constrained fit",{"type":25,"tag":243,"props":494,"children":495},{},[496,504],{"type":25,"tag":270,"props":497,"children":498},{},[499],{"type":25,"tag":40,"props":500,"children":501},{},[502],{"type":31,"value":503},"R²",{"type":25,"tag":270,"props":505,"children":506},{},[507],{"type":31,"value":508},"Goodness of fit — values > 0.99 are typical for clean dose-response data",{"type":25,"tag":243,"props":510,"children":511},{},[512,520],{"type":25,"tag":270,"props":513,"children":514},{},[515],{"type":25,"tag":40,"props":516,"children":517},{},[518],{"type":31,"value":519},"Residual plot",{"type":25,"tag":270,"props":521,"children":522},{},[523],{"type":31,"value":524},"Check for systematic pattern — random scatter confirms model is appropriate",{"type":25,"tag":243,"props":526,"children":527},{},[528,536],{"type":25,"tag":270,"props":529,"children":530},{},[531],{"type":25,"tag":40,"props":532,"children":533},{},[534],{"type":31,"value":535},"Steep sigmoid (large n)",{"type":25,"tag":270,"props":537,"children":538},{},[539],{"type":31,"value":540},"Cooperative system or a threshold-like switch",{"type":25,"tag":243,"props":542,"children":543},{},[544,552],{"type":25,"tag":270,"props":545,"children":546},{},[547],{"type":25,"tag":40,"props":548,"children":549},{},[550],{"type":31,"value":551},"Flat sigmoid (small n)",{"type":25,"tag":270,"props":553,"children":554},{},[555],{"type":31,"value":556},"Heterogeneous binding sites or measurement at narrow concentration range",{"type":25,"tag":26,"props":558,"children":560},{"id":559},"example-prompts",[561],{"type":31,"value":562},"Example Prompts",{"type":25,"tag":235,"props":564,"children":565},{},[566,582],{"type":25,"tag":239,"props":567,"children":568},{},[569],{"type":25,"tag":243,"props":570,"children":571},{},[572,577],{"type":25,"tag":247,"props":573,"children":574},{},[575],{"type":31,"value":576},"Scenario",{"type":25,"tag":247,"props":578,"children":579},{},[580],{"type":31,"value":581},"What to type",{"type":25,"tag":263,"props":583,"children":584},{},[585,602,619,636,653,670],{"type":25,"tag":243,"props":586,"children":587},{},[588,593],{"type":25,"tag":270,"props":589,"children":590},{},[591],{"type":31,"value":592},"Basic fit",{"type":25,"tag":270,"props":594,"children":595},{},[596],{"type":25,"tag":274,"props":597,"children":599},{"className":598},[],[600],{"type":31,"value":601},"fit Hill equation, concentration column is 'dose_uM', response is 'inhibition_pct'; report IC50 with 95% CI",{"type":25,"tag":243,"props":603,"children":604},{},[605,610],{"type":25,"tag":270,"props":606,"children":607},{},[608],{"type":31,"value":609},"Multiple compounds",{"type":25,"tag":270,"props":611,"children":612},{},[613],{"type":25,"tag":274,"props":614,"children":616},{"className":615},[],[617],{"type":31,"value":618},"fit Hill curves for each compound in the 'drug' column; overlay on one plot; rank by IC50",{"type":25,"tag":243,"props":620,"children":621},{},[622,627],{"type":25,"tag":270,"props":623,"children":624},{},[625],{"type":31,"value":626},"Cooperative binding",{"type":25,"tag":270,"props":628,"children":629},{},[630],{"type":25,"tag":274,"props":631,"children":633},{"className":632},[],[634],{"type":31,"value":635},"Hill equation fit with Hill coefficient; annotate EC50 and n; is n significantly different from 1?",{"type":25,"tag":243,"props":637,"children":638},{},[639,644],{"type":25,"tag":270,"props":640,"children":641},{},[642],{"type":31,"value":643},"Log-scale plot",{"type":25,"tag":270,"props":645,"children":646},{},[647],{"type":25,"tag":274,"props":648,"children":650},{"className":649},[],[651],{"type":31,"value":652},"sigmoidal dose-response fit on log10 concentration axis; add EC50 and 90% effect markers",{"type":25,"tag":243,"props":654,"children":655},{},[656,661],{"type":25,"tag":270,"props":657,"children":658},{},[659],{"type":31,"value":660},"Normalized data",{"type":25,"tag":270,"props":662,"children":663},{},[664],{"type":25,"tag":274,"props":665,"children":667},{"className":666},[],[668],{"type":31,"value":669},"normalize response 0–100% then fit Hill equation; report EC50, n, and goodness of fit",{"type":25,"tag":243,"props":671,"children":672},{},[673,678],{"type":25,"tag":270,"props":674,"children":675},{},[676],{"type":31,"value":677},"Inhibition with baseline",{"type":25,"tag":270,"props":679,"children":680},{},[681],{"type":25,"tag":274,"props":682,"children":684},{"className":683},[],[685],{"type":31,"value":686},"fit 4-parameter Hill (bottom, top, IC50, n); concentrations 0.001–100 µM",{"type":25,"tag":26,"props":688,"children":690},{"id":689},"assumptions-to-check",[691],{"type":31,"value":692},"Assumptions to Check",{"type":25,"tag":694,"props":695,"children":696},"ul",{},[697,707,717,727,737],{"type":25,"tag":147,"props":698,"children":699},{},[700,705],{"type":25,"tag":40,"props":701,"children":702},{},[703],{"type":31,"value":704},"Monotone response",{"type":31,"value":706}," — the Hill equation assumes response strictly increases (or decreases) with concentration; non-monotone data (e.g. bell-shaped) requires a different model",{"type":25,"tag":147,"props":708,"children":709},{},[710,715],{"type":25,"tag":40,"props":711,"children":712},{},[713],{"type":31,"value":714},"Wide enough concentration range",{"type":31,"value":716}," — the dataset should span from near-baseline to near-Emax; fitting only the steep middle region cannot reliably estimate EC50 or Emax",{"type":25,"tag":147,"props":718,"children":719},{},[720,725],{"type":25,"tag":40,"props":721,"children":722},{},[723],{"type":31,"value":724},"Replicate measurements",{"type":31,"value":726}," — single measurements at each concentration have no internal error estimate; include at least n=3 technical replicates per concentration",{"type":25,"tag":147,"props":728,"children":729},{},[730,735],{"type":25,"tag":40,"props":731,"children":732},{},[733],{"type":31,"value":734},"Appropriate baseline",{"type":31,"value":736}," — if the response does not start near 0 at low concentrations (e.g. partial agonist with non-zero floor), use the 4-parameter Hill equation with a free bottom parameter",{"type":25,"tag":147,"props":738,"children":739},{},[740,745],{"type":25,"tag":40,"props":741,"children":742},{},[743],{"type":31,"value":744},"Independence",{"type":31,"value":746}," — if the same sample is remeasured, observations are not independent; treat as technical replicates, not independent data points",{"type":25,"tag":26,"props":748,"children":750},{"id":749},"related-tools",[751],{"type":31,"value":752},"Related Tools",{"type":25,"tag":34,"props":754,"children":755},{},[756,758,764,766,772,774,780],{"type":31,"value":757},"Use the ",{"type":25,"tag":212,"props":759,"children":761},{"href":760},"/tools/michaelis-menten-fit",[762],{"type":31,"value":763},"Michaelis-Menten Fit",{"type":31,"value":765}," when your data follows a simple hyperbolic (non-sigmoidal) saturation model — enzyme velocity vs. substrate concentration without cooperativity. Use the ",{"type":25,"tag":212,"props":767,"children":769},{"href":768},"/tools/lineweaver-burk-plot",[770],{"type":31,"value":771},"Lineweaver-Burk Plot",{"type":31,"value":773}," to linearize enzyme kinetics data and identify inhibition type. Use the ",{"type":25,"tag":212,"props":775,"children":777},{"href":776},"/tools/logistic-regression",[778],{"type":31,"value":779},"Logistic Regression tool",{"type":31,"value":781}," when your outcome is binary (event/no event) rather than a continuous graded response.",{"type":25,"tag":26,"props":783,"children":785},{"id":784},"frequently-asked-questions",[786],{"type":31,"value":787},"Frequently Asked Questions",{"type":25,"tag":34,"props":789,"children":790},{},[791,796,800,802,806],{"type":25,"tag":40,"props":792,"children":793},{},[794],{"type":31,"value":795},"What is the difference between EC50 and IC50?",{"type":25,"tag":40,"props":797,"children":798},{},[799],{"type":31,"value":79},{"type":31,"value":801}," (half-maximal effective concentration) applies to activation assays — the concentration that produces 50% of the maximum stimulation (e.g. receptor agonists, enzyme activators). ",{"type":25,"tag":40,"props":803,"children":804},{},[805],{"type":31,"value":86},{"type":31,"value":807}," (half-maximal inhibitory concentration) applies to inhibition assays — the concentration that reduces activity by 50% (e.g. enzyme inhibitors, cytotoxicity, viral plaque reduction). Both are extracted identically from a Hill equation fit; the naming just reflects the direction of the effect.",{"type":25,"tag":34,"props":809,"children":810},{},[811,816,818,823],{"type":25,"tag":40,"props":812,"children":813},{},[814],{"type":31,"value":815},"My Hill coefficient is much greater than 1 — is that biologically meaningful?",{"type":31,"value":817},"\nA Hill coefficient n > 1 indicates ",{"type":25,"tag":40,"props":819,"children":820},{},[821],{"type":31,"value":822},"positive cooperativity",{"type":31,"value":824}," — binding of the first ligand increases affinity for subsequent ones. Classic example: hemoglobin (n ≈ 2.8). However, in pharmacological dose-response curves, n > 2 often reflects a steep but not mechanistically cooperative response — the Hill coefficient then serves as an empirical steepness descriptor rather than a literal measure of binding sites.",{"type":25,"tag":34,"props":826,"children":827},{},[828,833,835,840],{"type":25,"tag":40,"props":829,"children":830},{},[831],{"type":31,"value":832},"How do I compare IC50 values across compounds statistically?",{"type":31,"value":834},"\nFit each compound's curve independently to get IC50 and its standard error, then compare with a t-test on log(IC50) values (since IC50 follows a log-normal distribution). Alternatively, fit all curves simultaneously in a global model and test whether the IC50 parameters are equal. Ask the AI to ",{"type":25,"tag":196,"props":836,"children":837},{},[838],{"type":31,"value":839},"\"fit Hill curves for each compound and report IC50 with 95% CI for pairwise comparison\"",{"type":31,"value":841},".",{"type":25,"tag":34,"props":843,"children":844},{},[845,850,852,857,859,874],{"type":25,"tag":40,"props":846,"children":847},{},[848],{"type":31,"value":849},"What is the 4-parameter Hill equation and when do I need it?",{"type":31,"value":851},"\nThe standard (3-parameter) Hill equation assumes the response starts at exactly 0 at zero concentration and reaches exactly Emax at high concentration. The ",{"type":25,"tag":40,"props":853,"children":854},{},[855],{"type":31,"value":856},"4-parameter logistic (4PL)",{"type":31,"value":858}," adds a free bottom parameter: ",{"type":25,"tag":40,"props":860,"children":861},{},[862,864,868,869,873],{"type":31,"value":863},"Response = Bottom + (Top − Bottom) × ",{"type":25,"tag":53,"props":865,"children":866},{},[867],{"type":31,"value":57},{"type":31,"value":59},{"type":25,"tag":53,"props":870,"children":871},{},[872],{"type":31,"value":57},{"type":31,"value":65},{"type":31,"value":875},". Use 4PL when your data has a non-zero baseline (e.g. partial agonist, background fluorescence) or does not reach 100% inhibition.",{"type":25,"tag":34,"props":877,"children":878},{},[879,884],{"type":25,"tag":40,"props":880,"children":881},{},[882],{"type":31,"value":883},"My curve looks sigmoidal but R² is low — what is wrong?",{"type":31,"value":885},"\nLow R² with a visually good fit usually indicates high measurement noise or only 1–2 replicates. Check the residual plot — if residuals are random and small relative to the response range, the fit is fine despite low R². Low R² with a systematic residual pattern (S-shaped residuals) means the Hill equation is the wrong model — try a biphasic Hill equation or a Gaussian peak model.",{"title":7,"searchDepth":887,"depth":887,"links":888},2,[889,890,891,892,893,894,895,896],{"id":28,"depth":887,"text":32},{"id":138,"depth":887,"text":141},{"id":230,"depth":887,"text":233},{"id":404,"depth":887,"text":407},{"id":559,"depth":887,"text":562},{"id":689,"depth":887,"text":692},{"id":749,"depth":887,"text":752},{"id":784,"depth":887,"text":787},"markdown","content:tools:035.hill-equation-fit.md","content","tools/035.hill-equation-fit.md","tools/035.hill-equation-fit","md",{"loc":4},1775502468196]