31:T85c,possible_links = { let data = { "nodes": chain_nodes, "links" : chain_links }; let color = "blue" let height = 300; let width = 800; const links = data.links.map(d => Object.create(d)); const nodes = data.nodes.map(d => Object.create(d)); const simulation = d3.forceSimulation(nodes) .force("link", d3.forceLink(links).id(d => d.id).distance(d=> 100)) .force("charge", d3.forceManyBody().strength(-300)) .force("center", d3.forceCenter(width / 2, height / 2)); // Drag behavior defined const drag = simulation => { function dragstarted(event, d) { if (!event.active) simulation.alphaTarget(0.3).restart(); d.fx = d.x; d.fy = d.y; } function dragged(event, d) { d.fx = event.x; d.fy = event.y; } function dragended(event, d) { if (!event.active) simulation.alphaTarget(0); d.fx = null; d.fy = null; } return d3.drag() .on("start", dragstarted) .on("drag", dragged) .on("end", dragended); }; const svg = d3.select(DOM.svg(width, height)); const link = svg.append("g") .attr("stroke", "#999") .attr("stroke-opacity", 0.6) .selectAll("line") .data(links) .join("line") .attr("stroke-width", d => Math.sqrt(d.value)); const node = svg.append("g") .attr("stroke", "#fff") .attr("stroke-width", 1.5) .selectAll(".node") .data(nodes) .join("g") .attr('class','node') .attr("transform", d => {return "translate("+d.x+","+d.y+")"}) .attr("fill", color) .call(drag(simulation)); node.append("circle") .attr("r", 15) .attr("fill", color) node.append("text") .attr("dx", -5) .attr("dy", ".35em") .text(d => d.id); node.append("title") .text(d => d.id); simulation.on("tick", () => { link .attr("x1", d => d.source.x) .attr("y1", d => d.source.y) .attr("x2", d => d.target.x) .attr("y2", d => d.target.y); node .attr("transform", d => {return "translate("+d.x+","+d.y+")"}) }); invalidation.then(() => simulation.stop()); return svg.node(); }32:T51b,simDataOverview = md`First, simple transmission chains were simulated to ascertain performance under the most controlled of circumstances. In Figure X, we see an example of the workflow for two simulations. In this case, simulating a transmission chain of *8* with uniform distance between transmissions and uniform sampling time. A brief description of the workflow is as follows: 1. *matrix_for_BF* - A transmission chain with time between infections and respective sampling time is generated given parameters. 2. *seq_gen* - Using HyPhy and the supplied **SimulateSequence.bf** batch file, a fasta file is generated with simulated sequences from the matrix in the aforementioned step. 3. HIVTRACE is run on the dataset without any filtering, with edge filtering turned on for only triangles, and with both filtering for both triangles and cycles of size **${tex`{4}`}**. 4. Summaries and figures for each simulation is generated and then merged. A Snakefile is available that can be configured to adjust the transmission chain, number of samples, the distance between transmission represented as ${tex`{t\mu}`} from source to target, and the amount of within-host evolution that is accrued before the sequence is sampled, or ${tex`{t\mu}`} where ${tex`{t}`} is time between infection and sampling.` 33:T4c0,// Scatter plot of all edge differences linkWOFilterPlot = { const vals = const_tip_lengths[tip_length_key]; const marks = _.flatten(_.map(vals, v => { return _.zip(v['num-edges-dist'], _.fill(Array(v['num-edges-dist'].length), v.internal_length)) })); let m = _.unzip(marks); let x = m[1]; let y = m[0]; let hi = _.map(marks, d => { return {x:Number(d[1]), y: Number(d[0])} }) const regression = d3Regression.regressionLinear() .x(d => d.x) .y(d => d.y)(hi) var trace1 = { x: x, y: y, name: 'Total Links', mode: 'markers', type: 'scatter' }; let xr = _.map(_.range(1000), d => {return ((regression[1][0] - regression[0][0])/1000) * d}); let yr = _.map(xr, regression.predict); //debugger; var trace2 = { x: xr, y: yr, name: 'Regression', mode: 'lines' }; const div = DOM.element('div'); var layout = { title: "Links without Filtering (Uniform)", grid: {rows: 1, columns: 1, pattern: 'independent'}, xaxis: { title: "Distance between Transmissions" }, yaxis: { title: "Total Links" } }; const data = [trace1, trace2]; plotly.newPlot(div, data, layout); return div }34:Tbf5,tableMapper = d => { let t = _.pick(d, ['num_nodes', "avg-num-edges","avg-edges-removed", "avg-spurious","avg-wrongfully-purged", "avg-true-positives", "avg-false-positives", "avg-true-negatives", "avg-false-negatives"]); t["internal_length"] = d["internal_length"]; t["tip_length"] = d["tip_length"]; // Convert string representations of numbers to floats let tp = parseFloat(t["avg-true-positives"]); let tn = parseFloat(t["avg-true-negatives"]); let fp = parseFloat(t["avg-false-positives"]); let fn = parseFloat(t["avg-false-negatives"]); let totalEdges = parseFloat(t["avg-num-edges"]); let edgesRemoved = parseFloat(t["avg-edges-removed"]); // Calculations for confusion matrix metrics let precision = tp / (tp + fp); let recall = tp / (tp + fn); let specificity = tn / (tn + fp); let f1Score = 2 * (precision * recall) / (precision + recall); let mccNumerator = (tp * tn) - (fp * fn); let mccDenominator = Math.sqrt((tp + fp) * (tp + fn) * (tn + fp) * (tn + fn)); let mcc = mccNumerator / mccDenominator; // Calculating the metrics for scoring let effectiveness = tn / (tn + fp); let efficiency = tp / (tp + fn); let removalRatio = edgesRemoved / totalEdges; // Normalize scoring components to avoid division by zero effectiveness = isFinite(effectiveness) ? effectiveness : 0; efficiency = isFinite(efficiency) ? efficiency : 0; removalRatio = isFinite(removalRatio) ? removalRatio : 0; let w1 = 1/6; let w2 = 3/6; let w3 = 1/3; // Calculate the score let score = w1 * effectiveness + w2 * efficiency + w3 * removalRatio; // Formatting the score t["score"] = d3.format('.3f')(score); // Ensure finite numbers, replace NaN and Infinity with appropriate defaults precision = isFinite(precision) ? precision : 0; recall = isFinite(recall) ? recall : 0; specificity = isFinite(specificity) ? specificity : 0; f1Score = isFinite(f1Score) ? f1Score : 0; mcc = isFinite(mcc) ? mcc : 0; // Format the newly calculated metrics t["precision"] = d3.format('.3f')(precision); t["recall"] = d3.format('.3f')(recall); t["specificity"] = d3.format('.3f')(specificity); t["f1Score"] = d3.format('.3f')(f1Score); t["mcc"] = d3.format('.3f')(mcc); // Calculate accuracy (correcting the formula) t["accuracy"] = d3.format('.3f')((tp + tn) / (tp + tn + fp + fn)); // Format the existing metrics for consistency t["avg-edges-removed"] = d3.format('.3f')(t["avg-edges-removed"]); t["avg-spurious"] = d3.format('.3f')(t["avg-spurious"]); t["avg-wrongfully-purged"] = d3.format('.3f')(t["avg-wrongfully-purged"]); t["avg-true-positives"] = d3.format('.3f')(tp); t["avg-false-positives"] = d3.format('.3f')(fp); t["avg-true-negatives"] = d3.format('.3f')(tn); t["avg-false-negatives"] = d3.format('.3f')(fn); return t }35:T62a,### Table Description - **Inter-host Evolution Rate**: The average rate of evolutionary changes between different hosts. - **Intra-host Evolution Rate**: The average rate of evolutionary changes within the same host. - **Avg. Total Edges**: The average number of edges before filtering. - **Avg. True Positives**: The average number of edges that were correctly identified and retained by the filter. - **Avg. True Negatives**: The average number of edges that were correctly identified and removed by the filter. - **Avg. False Positives**: The average number of edges that were incorrectly retained, i.e., edges that should have been filtered out but were not. - **Avg. False Negatives**: The average number of edges that were incorrectly filtered out, i.e., edges that should have been retained but were removed. - **Network Filtering Score**: A composite metric calculated by weighting and averaging three key performance metrics: - **Effectiveness** (Weight: 1/6): The proportion of true negatives among edges removed, calculated as `(TN / (TN + FP))`. - **Efficiency** (Weight: 1/2): The proportion of true positives among edges retained, calculated as `(TP / (TP + FN))`. - **Edge Removal Ratio** (Weight: 1/3): The proportion of total edges removed, calculated as `((Total Edges - Edges After Filtering) / Total Edges)`. - This score evaluates the filtering performance by balancing the accuracy of edge removal, the precision in retaining correct edges, and the proportion of edges removed, allowing consistent comparison across various network configurations.36:Tde0,distanceHistogramBarChart = { const beta_dist = [2.4755131792324097, 2.761341300361401, 3.059823244203146, 3.3707427078967998, 3.6939017285596707, 4.029118408859609, 4.376225033419195, 4.73506649341659, 5.105498957272793, 5.487388740054944, 5.880611334988088, 6.28505057844602, 6.700597925786626, 7.127151819958805, 7.564617138318053, 8.012904705818215, 8.471930864890421, 8.94161709401894, 9.421889668380993, 9.912679357010203, 10.413921151829932, 10.925554024626239, 11.44752070862568, 11.97976750183432, 12.522244089704078, 13.074903385034085, 13.637701383302414, 14.210597031866277, 14.793552111674883, 15.386531130314768, 15.98950122535777, 16.602432077111686, 17.22529582998449, 17.85806702177117, 18.50072252025562, 19.153241466594434, 19.815605225014515, 20.48779733841335, 21.16980348950167, 21.86161146717383, 22.563211137831246, 23.274594421421114, 23.99575527198525, 24.726689662544544, 25.467395574171288, 26.21787298912777, 26.978123887972576, 27.74815225055878, 28.52796406086836, 29.317567315647786, 30.116972036828415, 30.926190287733935, 31.74523619309541, 32.5741259629123, 33.41287792021549, 34.261512532806776, 35.120052449066925, 35.98852253794363, 36.866949933249124, 37.755364082418076, 38.653796799896256, 39.56228232535303, 40.48085738693355, 41.40956126979114, 42.348435890166655, 43.29752587530976, 44.256878649567334, 45.22654452699697, 46.20657681089871, 47.19703190069743, 48.19796940664938, 49.209452272892605, 50.23154690941171, 51.264323333541284, 52.30785532169448, 53.36222057206845, 54.42750087915295, 55.503782320949774, 56.59115545990034, 57.689715558619625, 58.799562811645536, 59.92080259453685, 61.05354573179116, 62.19790878520855, 63.35401436450062, 64.52199146213715, 65.7019758146418, 66.89411029279329, 68.09854532346468, 69.31543934614723, 70.54495930755793, 71.7872811981324, 73.04259063466145, 74.31108349384824, 75.59296660215945, 76.88845848802063, 78.19779020318819, 79.52120622102491, 80.8589654204409, 82.21134216545545, 83.57862749172023, 84.96113041295307, 86.35917936210723, 87.77312378429319, 89.20333590104174, 90.65021266852241, 92.11417795590398, 93.59568497427517, 95.0952189915769, 96.6133003750071, 98.15048800955873, 99.70738315002309, 101.28463377427038, 102.88293951835131, 104.50305728950313, 106.1458076722093, 107.81208226597737, 109.50285212267424, 111.21917748766374, 112.96221909471645, 114.73325132247166, 116.53367759385243, 118.36504849428545, 120.22908320675482, 122.12769502110217, 124.06302188483815, 126.03746324176642, 128.05372477967938, 130.1148732180406, 132.2244039678636, 134.3863254739863, 136.60526543373146, 138.88660607523107, 141.23665858860417, 143.66289114214987, 146.1742315243818, 148.78147575923302, 151.49785054709693, 154.33980462044735, 157.32815060577633, 160.48976156099386, 163.8601798455159, 167.4877970640112, 171.44089387296236, 175.82025555125787, 180.78365075361026, 186.59864292900676, 193.77494559720304, 203.48289797536182, 219.59521455316255] const no_filter = { x: _.map(lanl_distances, d=> d.Distance), type: 'histogram', name: 'No Filtering', xbins: {"start": 0.0, "end": 0.015}, nbinsx : 20 }; const beta_line = { x: _.map(beta_dist, (d, i) => i*.0001), y: _.map(beta_dist, d=> d * 2000), type: 'line', name: 'Beta', line: { dash: 'dashdot', width: 1 } }; const data = [no_filter, beta_line]; const div = DOM.element('div'); var layout = { title: "Distance Histogram", xaxis: { title: 'Distance' } }; plotly.newPlot(div, data, layout); return div }37:T400,favitesTable = table(_.map(favites_report, d => { let t = _.pick(d, ['num_nodes', "num_edges","avg-edges-removed", "avg-spurious","avg-wrongfully-purged", "avg-cycles-removed", "avg-cycle-spurious","avg-cycle-wrongfully-purged"]); //t["avg_total_edges"] = d["avg-num-edges"] + ["avg-edges-removed"]; t["avg-edges-removed"] = text_float_format(t["avg-edges-removed"]); t["avg-spurious"] = text_float_format(t["avg-spurious"]); t["avg-wrongfully-purged"] = text_float_format(t["avg-wrongfully-purged"]); t["avg-cycles-removed"] = text_float_format(t["avg-cycles-removed"]); t["avg-cycle-spurious"] = text_float_format(t["avg-cycle-spurious"]); t["avg-cycle-wrongfully-purged"] = text_float_format(t["avg-cycle-wrongfully-purged"]); t["FNR"] = text_float_format(d["avg-spurious"]/(d["avg-spurious"] + d["avg-edges-removed"])) t['FNR-cycle'] = text_float_format(d["avg-cycle-spurious"]/(d["avg-cycle-spurious"] + d["avg-cycles-removed"])) return t }) )38:T8a7,DSPSBarChart = { const no_filter = { x: _.range(dsps_no_filter["trace_results"]["Degrees"]["Distribution"].length), y: dsps_no_filter["trace_results"]["Degrees"]["Distribution"], type: 'bar', name: 'No Filtering' }; const nf_rho = { x: _.range(dsps_no_filter["trace_results"]["Degrees"]["Distribution"].length), y: _.map(dsps_no_filter["trace_results"]["Degrees"]["fitted"], d => {return 1 + d * d3.sum(dsps_no_filter["trace_results"]["Degrees"]["Distribution"])}), type: 'line', showlegend: false, line: { dash: 'dashdot', width: 1 } }; const filter = { x: _.range(dsps_filter["trace_results"]["Degrees"]["Distribution"].length), y: dsps_filter["trace_results"]["Degrees"]["Distribution"], xaxis: 'x2', yaxis: 'y2', type: 'bar', name: 'C₃ Edge Filtering' }; const f_rho = { x: _.range(dsps_filter["trace_results"]["Degrees"]["Distribution"].length), y: _.map(dsps_filter["trace_results"]["Degrees"]["fitted"], d => {return 1 + d * d3.sum(dsps_filter["trace_results"]["Degrees"]["Distribution"])}), type: 'line', xaxis: 'x2', yaxis: 'y2', showlegend: false, line: { dash: 'dashdot', width: 1 } }; const cycle_filter = { x: _.range(dsps_cycle_filter["trace_results"]["Degrees"]["Distribution"].length), y: dsps_cycle_filter["trace_results"]["Degrees"]["Distribution"], xaxis: 'x3', yaxis: 'y3', type: 'bar', name: 'C_3,4 Edge Filtering' }; const cycle_rho = { x: _.range(dsps_cycle_filter["trace_results"]["Degrees"]["Distribution"].length), y: _.map(dsps_cycle_filter["trace_results"]["Degrees"]["fitted"], d => {return 1 + d * d3.sum(dsps_cycle_filter["trace_results"]["Degrees"]["Distribution"])}), type: 'line', xaxis: 'x3', yaxis: 'y3', showlegend: false, line: { dash: 'dashdot', width: 1 } }; const data = [no_filter, nf_rho, filter, f_rho, cycle_filter, cycle_rho]; const div = DOM.element('div'); var layout = { title: "Degree Distribution", grid: {rows: 1, columns: 3, pattern: 'independent'}, xaxis: { title: 'Degree' } }; plotly.newPlot(div, data, layout); return div } 39:T1570,autotuneEdgeRemovalPlot = { const no_filter = { x: _.range(autotune_datasets[datasetIDKey]["nofilter_results"]["trace_results"]["Degrees"]["Distribution"].length), y: autotune_datasets[datasetIDKey]["nofilter_results"]["trace_results"]["Degrees"]["Distribution"], type: 'bar', name: 'No Filtering' }; const nf_rho = { x: _.range(autotune_datasets[datasetIDKey]["nofilter_results"]["trace_results"]["Degrees"]["Distribution"].length), y: _.map(autotune_datasets[datasetIDKey]["nofilter_results"]["trace_results"]["Degrees"]["fitted"], d => {return 1 + d * d3.sum(autotune_datasets[datasetIDKey]["nofilter_results"]["trace_results"]["Degrees"]["Distribution"])}), type: 'line', showlegend: false, line: { dash: 'dashdot', width: 1 } }; const df_no_filter = { x: _.range(autotune_datasets[datasetIDKey]["default_nofilter_results"]["trace_results"]["Degrees"]["Distribution"].length), y: autotune_datasets[datasetIDKey]["default_nofilter_results"]["trace_results"]["Degrees"]["Distribution"], xaxis: 'x4', yaxis: 'y4', type: 'bar', name: 'No Filtering @ 1.5%' }; const dnf_rho = { x: _.range(autotune_datasets[datasetIDKey]["default_nofilter_results"]["trace_results"]["Degrees"]["Distribution"].length), y: _.map(autotune_datasets[datasetIDKey]["default_nofilter_results"]["trace_results"]["Degrees"]["fitted"], d => {return 1 + d * d3.sum(autotune_datasets[datasetIDKey]["default_nofilter_results"]["trace_results"]["Degrees"]["Distribution"])}), xaxis: 'x4', yaxis: 'y4', type: 'line', showlegend: false, line: { dash: 'dashdot', width: 1 } }; const filter = { x: _.range(autotune_datasets[datasetIDKey]["filter_results"]["trace_results"]["Degrees"]["Distribution"].length), y: autotune_datasets[datasetIDKey]["filter_results"]["trace_results"]["Degrees"]["Distribution"], xaxis: 'x2', yaxis: 'y2', type: 'bar', name: 'C₃ Edge Filtering' }; const f_rho = { x: _.range(autotune_datasets[datasetIDKey]["filter_results"]["trace_results"]["Degrees"]["Distribution"].length), y: _.map(autotune_datasets[datasetIDKey]["filter_results"]["trace_results"]["Degrees"]["fitted"], d => {return 1 + d * d3.sum(autotune_datasets[datasetIDKey]["filter_results"]["trace_results"]["Degrees"]["Distribution"])}), type: 'line', xaxis: 'x2', yaxis: 'y2', showlegend: false, line: { dash: 'dashdot', width: 1 } }; const default_filter = { x: _.range(autotune_datasets[datasetIDKey]["default_filter_results"]["trace_results"]["Degrees"]["Distribution"].length), y: autotune_datasets[datasetIDKey]["default_filter_results"]["trace_results"]["Degrees"]["Distribution"], xaxis: 'x5', yaxis: 'y5', type: 'bar', name: 'C₃ Edge Filtering @ 1.5%' }; const df_rho = { x: _.range(autotune_datasets[datasetIDKey]["default_filter_results"]["trace_results"]["Degrees"]["Distribution"].length), y: _.map(autotune_datasets[datasetIDKey]["default_filter_results"]["trace_results"]["Degrees"]["fitted"], d => {return 1 + d * d3.sum(autotune_datasets[datasetIDKey]["default_filter_results"]["trace_results"]["Degrees"]["Distribution"])}), type: 'line', xaxis: 'x5', yaxis: 'y5', showlegend: false, line: { dash: 'dashdot', width: 1 } }; const cycle_filter = { x: _.range(autotune_datasets[datasetIDKey]["cycle_results"]["trace_results"]["Degrees"]["Distribution"].length), y: autotune_datasets[datasetIDKey]["cycle_results"]["trace_results"]["Degrees"]["Distribution"], xaxis: 'x3', yaxis: 'y3', type: 'bar', name: 'C_3,4 Edge Filtering' }; const cycle_rho = { x: _.range(autotune_datasets[datasetIDKey]["cycle_results"]["trace_results"]["Degrees"]["Distribution"].length), y: _.map(autotune_datasets[datasetIDKey]["cycle_results"]["trace_results"]["Degrees"]["fitted"], d => {return 1 + d * d3.sum(autotune_datasets[datasetIDKey]["cycle_results"]["trace_results"]["Degrees"]["Distribution"])}), type: 'line', xaxis: 'x3', yaxis: 'y3', showlegend: false, line: { dash: 'dashdot', width: 1 } }; const default_cycle_filter = { x: _.range(autotune_datasets[datasetIDKey]["default_cycle_results"]["trace_results"]["Degrees"]["Distribution"].length), y: autotune_datasets[datasetIDKey]["default_cycle_results"]["trace_results"]["Degrees"]["Distribution"], xaxis: 'x6', yaxis: 'y6', type: 'bar', name: 'C_3,4 Edge Filtering @ 1.5%' }; const dcycle_rho = { x: _.range(autotune_datasets[datasetIDKey]["default_cycle_results"]["trace_results"]["Degrees"]["Distribution"].length), y: _.map(autotune_datasets[datasetIDKey]["default_cycle_results"]["trace_results"]["Degrees"]["fitted"], d => {return 1 + d * d3.sum(autotune_datasets[datasetIDKey]["default_cycle_results"]["trace_results"]["Degrees"]["Distribution"])}), type: 'line', xaxis: 'x6', yaxis: 'y6', showlegend: false, line: { dash: 'dashdot', width: 1 } }; const data = [no_filter, nf_rho, filter, f_rho, cycle_filter, cycle_rho, df_no_filter, dnf_rho, default_filter, df_rho, default_cycle_filter, dcycle_rho]; const div = DOM.element('div'); var layout = { title: "Degree Distribution", grid: {rows: 2, columns: 3, pattern: 'independent'}, xaxis: { title: 'Degree' } }; plotly.newPlot(div, data, layout); return div }3a:T414,bics = _.map(autotune_datasets, (d, key) => {return { "id": d["author"], "type" : "No Filtering", "bic" : d["nofilter_results"]["trace_results"]["Degrees"]["BIC"]}}).concat(_.map(autotune_datasets, (d, key) => {return { "id": d["author"], "type" : "C3 Edge Filtering", "bic" : d["filter_results"]["trace_results"]["Degrees"]["BIC"]}})).concat(_.map(autotune_datasets, (d, key) => {return { "id": d["author"], "type" : "C3,4 Edge Filtering", "bic" : d["cycle_results"]["trace_results"]["Degrees"]["BIC"]}})).concat(_.map(autotune_datasets, (d, key) => {return { "id": d["author"], "type" : "1.5% No Filtering", "bic" : d["default_nofilter_results"]["trace_results"]["Degrees"]["BIC"]}})).concat(_.map(autotune_datasets, (d, key) => {return { "id": d["author"], "type" : "1.5% C3 Filtering", "bic" : d["default_filter_results"]["trace_results"]["Degrees"]["BIC"]}})).concat(_.map(autotune_datasets, (d, key) => {return { "id": d["author"], "type" : "1.5% C3,C4 Filtering", "bic" : d["default_cycle_results"]["trace_results"]["Degrees"]["BIC"]}}))3b:T89d,nccEdgeRemovalPlot = { const no_filter = { x: _.range(ncc_no_filter["trace_results"]["Degrees"]["Distribution"].length), y: ncc_no_filter["trace_results"]["Degrees"]["Distribution"], type: 'bar', name: 'No Filtering' }; const nf_rho = { x: _.range(ncc_no_filter["trace_results"]["Degrees"]["Distribution"].length), y: _.map(ncc_no_filter["trace_results"]["Degrees"]["fitted"], d => {return 1 + d * d3.sum(ncc_no_filter["trace_results"]["Degrees"]["Distribution"])}), type: 'line', showlegend: false, line: { dash: 'dashdot', width: 1 } }; const filter = { x: _.range(ncc_filter["trace_results"]["Degrees"]["Distribution"].length), y: ncc_filter["trace_results"]["Degrees"]["Distribution"], xaxis: 'x2', yaxis: 'y2', type: 'bar', name: 'C₃ Edge Filtering' }; const f_rho = { x: _.range(ncc_filter["trace_results"]["Degrees"]["Distribution"].length), y: _.map(ncc_filter["trace_results"]["Degrees"]["fitted"], d => {return 1 + d * d3.sum(ncc_filter["trace_results"]["Degrees"]["Distribution"])}), type: 'line', xaxis: 'x2', yaxis: 'y2', showlegend: false, line: { dash: 'dashdot', width: 1 } }; const cycle_filter = { x: _.range(ncc_cycle_filter["trace_results"]["Degrees"]["Distribution"].length), y: ncc_cycle_filter["trace_results"]["Degrees"]["Distribution"], xaxis: 'x3', yaxis: 'y3', type: 'bar', name: 'C_3,4 Edge Filtering' }; const cycle_rho = { x: _.range(ncc_cycle_filter["trace_results"]["Degrees"]["Distribution"].length), y: _.map(ncc_cycle_filter["trace_results"]["Degrees"]["fitted"], d => {return 1 + d * d3.sum(ncc_cycle_filter["trace_results"]["Degrees"]["Distribution"])}), type: 'line', xaxis: 'x3', yaxis: 'y3', showlegend: false, line: { dash: 'dashdot', width: 1 } }; const data = [no_filter, nf_rho, filter, f_rho, cycle_filter, cycle_rho]; const div = DOM.element('div'); var layout = { title: "Degree Distribution", grid: {rows: 1, columns: 3, pattern: 'independent'}, xaxis: { title: 'Degree' } }; plotly.newPlot(div, data, layout); return div }3c:T4b3,lanlBarChart = { const no_filter = { x: _.range(lanl_no_filter["Distribution"].length), y: lanl_no_filter["Distribution"], type: 'bar', name: 'No Filtering' }; const filter = { x: _.range(lanl_filter["Distribution"].length), y: lanl_no_filter["Distribution"], xaxis: 'x2', yaxis: 'y2', type: 'bar', name: 'C₃ Edge Filtering' }; const cycle_filter = { x: _.range(lanl_cycle_filter["Distribution"].length), y: lanl_cycle_filter["Distribution"], xaxis: 'x3', yaxis: 'y3', type: 'bar', name: 'C_3,4 Edge Filtering' }; const data = [no_filter, filter, cycle_filter]; const div = DOM.element('div'); var layout = { title: "Degree Distribution - log-log plot", grid: {rows: 1, columns: 3, pattern: 'independent'}, xaxis: { type: 'log', title: 'Degree' }, yaxis: { type: 'log', title: 'Count' }, xaxis2: { type: 'log', title: 'Degree' }, yaxis2: { type: 'log', title: 'Count' }, xaxis3: { type: 'log', title: 'Degree' }, yaxis3: { type: 'log', title: 'Count' }, }; plotly.newPlot(div, data, layout); return div }3d:Tdfb,chain = { let data = { "nodes": [ {"id": "A", "group": 1}, {"id": "B", "group": 1}, {"id": "C", "group": 1} ], "links" : [ {"source": "A", "target": "B", "value": 30, "name" : "d1" }, {"source": "B", "target": "C", "value": 30, "name" : "d2"} ] }; let color = "blue" let height = 200; let width = 200; const links = data.links.map(d => Object.create(d)); const nodes = data.nodes.map(d => Object.create(d)); const simulation = d3.forceSimulation(nodes) .force("link", d3.forceLink(links).id(d => d.id).distance(d=> 80)) .force("charge", d3.forceManyBody()) .force("center", d3.forceCenter(width / 2, height / 2)); const svg = d3.select(DOM.svg(width, height)); svg.append('defs').append('marker') .attr("id","arrowhead") .attr('viewBox','-0 -5 10 10') .attr('refX',13) .attr('refY',0) .attr('orient','auto') .attr('markerWidth',3) .attr('markerHeight',3) .attr('xoverflow','visible') .append('svg:path') .attr('d', 'M 0,-5 L 5,0 L 0,5') .attr('fill', '#999') .style('stroke','none'); const link = svg.append("g") .attr("stroke", "#999") .attr("stroke-opacity", 0.6) .selectAll("line") .data(links) .join("line") .attr("stroke-width", d => Math.sqrt(d.value)) .attr('marker-end','url(#arrowhead)'); const edgepaths = svg.selectAll(".edgepath") .data(links) .enter() .append('path') .attr('class', 'edgepath') .attr('fill-opacity', 0) .attr('stroke-opacity', 0) .attr('id', function (d, i) {return 'edgepath' + i}) .style("pointer-events", "none"); const edgelabels = svg.selectAll(".edgelabel") .data(links) .enter() .append('text') .style("pointer-events", "none") .style("pointer-events", "none") .attr('class', 'edgelabel') .attr('id', (d, i) => {return 'edgelabel' + i}) .attr('font-size', 14) edgelabels.append('textPath') .attr('xlink:href', function (d, i) {return '#edgepath' + i}) .style("text-anchor", "middle") .style("pointer-events", "none") .attr("startOffset", "50%") .text(function (d) {return d.name}); const node = svg.append("g") .attr("stroke", "#fff") .attr("stroke-width", 1.5) .selectAll(".node") .data(nodes) .join("g") .attr('class','node') .attr("transform", d => {return "translate("+d.x+","+d.y+")"}) .attr("fill", color) .call(drag(simulation)); node.append("circle") .attr("r", 15) .attr("fill", color) node.append("text") .attr("dx", -5) .attr("dy", ".35em") .text(d => d.id); node.append("title") .text(d => d.id); simulation.on("tick", () => { link .attr("x1", d => d.source.x) .attr("y1", d => d.source.y) .attr("x2", d => d.target.x) .attr("y2", d => d.target.y); edgepaths.attr('d', function (d) { return 'M ' + d.source.x + ' ' + d.source.y + ' L ' + d.target.x + ' ' + d.target.y; }); edgelabels.attr('transform', function (d) { if (d.target.x < d.source.x) { let bbox = this.getBBox(); let rx = bbox.x + bbox.width / 2; let ry = bbox.y + bbox.height / 2; return 'rotate(180 ' + rx + ' ' + ry + ')'; } else { return 'rotate(0)'; } }); node .attr("transform", d => {return "translate("+d.x+","+d.y+")"}) }); invalidation.then(() => simulation.stop()); return svg.node(); }3e:T737,chart = { let data = { "nodes": [ {"id": "A", "group": 1}, {"id": "B", "group": 1}, {"id": "C", "group": 1}, ], "links" : [ {"source": "A", "target": "B", "value": 30}, {"source": "B", "target": "C", "value": 30}, {"source": "A", "target": "C", "value": 30}, ] }; let color = "blue" let height = 200; let width = 200; const links = data.links.map(d => Object.create(d)); const nodes = data.nodes.map(d => Object.create(d)); const simulation = d3.forceSimulation(nodes) .force("link", d3.forceLink(links).id(d => d.id).distance(d=> 80)) .force("charge", d3.forceManyBody()) .force("center", d3.forceCenter(width / 2, height / 2)); const svg = d3.select(DOM.svg(width, height)); const link = svg.append("g") .attr("stroke", "#999") .attr("stroke-opacity", 0.6) .selectAll("line") .data(links) .join("line") .attr("stroke-width", d => Math.sqrt(d.value)); const node = svg.append("g") .attr("stroke", "#fff") .attr("stroke-width", 1.5) .selectAll(".node") .data(nodes) .join("g") .attr('class','node') .attr("transform", d => {return "translate("+d.x+","+d.y+")"}) .attr("fill", color) .call(drag(simulation)); node.append("circle") .attr("r", 15) .attr("fill", color) node.append("text") .attr("dx", -5) .attr("dy", ".35em") .text(d => d.id); node.append("title") .text(d => d.id); simulation.on("tick", () => { link .attr("x1", d => d.source.x) .attr("y1", d => d.source.y) .attr("x2", d => d.target.x) .attr("y2", d => d.target.y); node .attr("transform", d => {return "translate("+d.x+","+d.y+")"}) }); invalidation.then(() => simulation.stop()); return svg.node(); }3f:T46b,autotuneBiccDiffs = _.map(autotune_datasets, d => { const nofilterBIC = d["nofilter_results"]["trace_results"]["Degrees"]["BIC"]; const filterBIC = d["filter_results"]["trace_results"]["Degrees"]["BIC"]; const cycleBIC = d["cycle_results"]["trace_results"]["Degrees"]["BIC"]; const defaultNofilterBIC = d["default_nofilter_results"]["trace_results"]["Degrees"]["BIC"]; const defaultFilterBIC = d["default_filter_results"]["trace_results"]["Degrees"]["BIC"]; const defaultCycleBIC = d["default_cycle_results"]["trace_results"]["Degrees"]["BIC"]; const filterDiff = nofilterBIC - filterBIC; const cycleDiff = nofilterBIC - cycleBIC; const filterCycleDiff = filterBIC - cycleBIC; const defaultFilterDiff = defaultNofilterBIC - defaultFilterBIC; const defaultCycleDiff = defaultNofilterBIC - defaultCycleBIC; const defaultFilterCycleDiff = defaultFilterBIC - defaultCycleBIC; // Calculate percentage differences using the nofilter BIC as base where applicable return { filterDiff, cycleDiff, filterCycleDiff, defaultFilterDiff, defaultCycleDiff, defaultFilterCycleDiff }; });40:T724,autotuneDataSetTables = { autotune_datasets[datasetIDKey]["nofilter_results"]["trace_results"]["Degrees"]["Type"] = "No Filtering" autotune_datasets[datasetIDKey]["filter_results"]["trace_results"]["Degrees"]["Type"] = "C₃ Edge Filtering" autotune_datasets[datasetIDKey]["cycle_results"]["trace_results"]["Degrees"]["Type"] = "C_3,4 Edge Filtering" autotune_datasets[datasetIDKey]["default_nofilter_results"]["trace_results"]["Degrees"]["Type"] = "No Filtering @ 1.5%" autotune_datasets[datasetIDKey]["default_filter_results"]["trace_results"]["Degrees"]["Type"] = "C₃ Edge Filtering @ 1.5%" autotune_datasets[datasetIDKey]["default_cycle_results"]["trace_results"]["Degrees"]["Type"] = "C_3,4 Edge Filtering @ 1.5%" var tables = [ autotune_datasets[datasetIDKey]["nofilter_results"]["trace_results"]["Degrees"], autotune_datasets[datasetIDKey]["filter_results"]["trace_results"]["Degrees"], autotune_datasets[datasetIDKey]["cycle_results"]["trace_results"]["Degrees"], autotune_datasets[datasetIDKey]["default_nofilter_results"]["trace_results"]["Degrees"], autotune_datasets[datasetIDKey]["default_filter_results"]["trace_results"]["Degrees"], autotune_datasets[datasetIDKey]["default_cycle_results"]["trace_results"]["Degrees"], ] tables = _.map(tables, t => _.pick(t, ["Type", "BIC", "rho", "rho CI"])) _.each(tables, t => { t["BIC"] = text_float_format(t["BIC"]); t["ρ"] = text_float_format(t["rho"]); t["ρ CI"] = interval_format(t["rho CI"]); delete t["rho"] delete t["rho CI"] }) return tables }41:T540,triangleDayDiffs = triangles.map(triangle => { const [a, b, c, value] = triangle; const edges = { [a + b]: [timeDiffLookup[a + b] || 'No data', filteredEdges.has(a + b)], [b + c]: [timeDiffLookup[b + c] || 'No data', filteredEdges.has(b + c)], [c + a]: [timeDiffLookup[c + a] || 'No data', filteredEdges.has(c + a)] }; // Initialize arrays to collect ages let filteredAges = []; let unfilteredAges = []; // Process each edge Object.entries(edges).forEach(([key, [age, isFiltered]]) => { if (isFiltered && age !== 'No data') { filteredAges.push(age); } else if (!isFiltered && age !== 'No data') { unfilteredAges.push(age); } }); // Calculate average ages const avgFilteredAge = filteredAges.length > 0 ? filteredAges.reduce((a, b) => a + b, 0) / filteredAges.length : 0; const avgUnfilteredAge = unfilteredAges.length > 0 ? unfilteredAges.reduce((a, b) => a + b, 0) / unfilteredAges.length : 0; // Annotate each triangle with the calculated averages return { triangle: [a, b, c], value: value, ...edges, avgFilteredAge: avgFilteredAge, avgUnfilteredAge: avgUnfilteredAge, filteredOlder: avgFilteredAge > avgUnfilteredAge }; }); 42:[{"id":"b1f0e0cccf34d9fd28921cf49f789030b31aa9d26030f7b9f4b09e44805c2431174965897450833910c0c95606028057f3e53464a0910b2de2fe9d062891b69c","href":"https://static.observableusercontent.com/files/b1f0e0cccf34d9fd28921cf49f789030b31aa9d26030f7b9f4b09e44805c2431174965897450833910c0c95606028057f3e53464a0910b2de2fe9d062891b69c","name":"betweenness-Table 1.csv","type":"text/csv","lastModified":1721413721317.579,"size":23732}] 2f:{"title":"Edge Filtering Report","cells":[{"id":0,"value":"intro = md`# Edge Filtering Report\n## Introduction\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":2496,"value":"$${chain} ${chart}","pinned":false,"hidden":false,"mode":"md"},{"id":101,"value":"overview = md`\n\n\nThe closer the distance between viral sequence data, the number of edges in the inferred network without filtering will continue to approach a fully connected network. To understand why this occurs, take into account a transmission chain between three individuals. When a threshold of ${tex`t`} is set to create a link between two nodes, and ${tex`{d_1 + d_2 <= t}`}, then a triangle will be inferred by HIV-TRACE.\n\n${chain} ${chart}\n** Figure 1 ** When ${tex`{d_1}`} and ${tex`{d_2}`} are small, a triangle will be inferred.\n\nThe maximum possible number of edges for ${tex`{N}`} nodes is ${tex`L_{max} = \\frac{N(N−1)}{2}`}. For example, a network of size ${tex`{N=8}`} has ${tex`L_{max} = 28`}. \n\nIn a chain of 8 nodes, there are ${tex`L = 7`} edges. When simulating transmission with a fixed rate across all chains, ${tex`d_{1} = d_{2} = d_{n} = r`}, given that the time of sampling is close to infection, then the number of spurious edges will follow a degree distribution like the following. \n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1715,"value":"example_header = md` ## Example`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1564,"value":"viewof mutation_rate = DOM.range(0.005, 0.025, .001)","pinned":false,"hidden":false,"mode":"ojs"},{"id":1725,"value":"mut_rate_explanation = md` *The **constant mutation rate** for this example is currently set to **${mutation_rate}**. Adjust the slider to set the constant mutation rate. In empirical and simulated datasets, distances between sequences are likely to be variable.*`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1577,"value":"viewof threshold = DOM.range(0.010, 0.050, .005)","pinned":false,"hidden":false,"mode":"ojs"},{"id":1737,"value":"threshold_explanation = md`*The **threshold**, set to **${threshold}**. Two sequences will be connected with a putative link, if and only if their pairwise distance does not exceed this threshold. In **HIV-TRACE**, the threshold is set by the -t or --threshold flag. Feel free to adjust this value.*`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1515,"value":"$31","pinned":false,"hidden":false,"mode":"ojs"},{"id":1743,"value":"possible_links_caption = md`*As the distances between sequences are set shorter, or the threshold is increased, the network approaches being fully connected. Out of a maximum number of links, **28**, and an expected **7** links, the total number of links with the set parameters are **${_.sum(distribution)/2}**. *`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1609,"value":"links_bar_chart = {\n \nconst no_filter = {\n x: _.map(chain_nodes, d=>d.id),\n y: distribution,\n type: 'bar',\n name: 'No Filtering'\n };\n\nconst data = [no_filter];\nconst div = DOM.element('div');\n\nvar layout = {\n title: \"Degree Distribution\",\n xaxis: {\n title: 'Node'\n }\n};\n \n plotly.newPlot(div, data, layout);\n return div\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":1880,"value":"links_bar_chart_caption = md`*The degree of each node in the network. When the threshold is set higher, the degree of each node is higher.*`","pinned":false,"hidden":false,"mode":"ojs"},{"id":2133,"value":"exampleDegreeTable = table(_.map(distribution, (d,i) => { return { Node: i, Degree : d}}))","pinned":false,"hidden":false,"mode":"ojs"},{"id":2148,"value":"exampleEdgeTable = table(_.map(chain_links, d=> _.omit(d, ['value'])))","pinned":false,"hidden":false,"mode":"ojs"},{"id":1595,"value":"simDataHeader = md`## Simulated Data`","pinned":false,"hidden":false,"mode":"ojs"},{"id":2079,"value":"$32","pinned":false,"hidden":false,"mode":"ojs"},{"id":2084,"value":"loadsubdag = {\n const svg = d3\n .create(\"svg\")\n .attr(\"viewBox\", [0, 0, layout.width, layout.height]);\n\n // fit to layout size\n const svgELScale = d3.select(subdag);\n svgELScale.attr('width', layout.width).attr('height', layout.height); // Just to zoom\n // append to SVG\n svg.node().appendChild(subdag);\n\n return svg.node();\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":2073,"value":"repCaption = md`#### _Averages from **20** repetitions with chain lengths of **8**_`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1258,"value":"viewof tip_length_key = DOM.range(0.001, 0.009, .001)","pinned":false,"hidden":false,"mode":"ojs"},{"id":1756,"value":"tipLengthLabelCaption = md `*With sampling length of **${tip_length_key}**.*`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1229,"value":"$33","pinned":false,"hidden":false,"mode":"ojs"},{"id":1226,"value":"linkWOFilterPlotCaption = md`**Figure 2** *Fewer edges are created as the fixed distances between hosts increase. The distance threshold used is **0.015** and infection to sampling intra-host distance of **${tip_length_key}**. Transmission chains simulated that are more closely related will stray further from recovering the transmission chain composed of 7 links, and closer to a fully connected graph, while increasing sampling distance from transmission reduces clustering.*`","pinned":false,"hidden":false,"mode":"ojs"},{"id":665,"value":"heatMapHeader = md`#### Heat Map`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1119,"value":"viewof heatmap_key = DOM.select(_.keys(key_labels))","pinned":false,"hidden":false,"mode":"ojs"},{"id":2463,"value":"heatMapViz = Plot.plot({\n x : {\n label: \"Simulated between host evolution (Internal node length)\"\n },\n y : {\n label: \"Simulated intra-host evolution (Tip length)\"\n },\n color: {\n scheme: \"turbo\",\n type: \"symlog\",\n legend: true\n },\n marks: [\n Plot.rect(_.filter(_.flatten(_.values(const_tip_lengths)), d=> d[heatmap_key] >= 0 ), Plot.bin({fillOpacity: \"count\"}, {x: \"internal_length\", y: \"tip_length\", fill: heatmap_key}))\n ]\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":141,"value":"edgeRemovalTableHeader = md`### Edge Removal Tables`","pinned":false,"hidden":false,"mode":"ojs"},{"id":153,"value":"edgeRemovalTable = Inputs.table(_.map(reports, tableMapper), edgeFilterHeaders)\n","pinned":false,"hidden":false,"mode":"ojs"},{"id":3326,"value":"[_.min(_.map(_.filter(reports, d => d[\"tip_length\"] == \"0.001\" && parseFloat(d[\"internal_length\"]) <= 0.01), d => { return d[\"avg-edges-removed\"]/(d[\"avg-num-edges\"] - 7) })), _.max(_.map(_.filter(reports, d => d[\"tip_length\"] == \"0.001\" && parseFloat(d[\"internal_length\"]) <= 0.01), d => { return d[\"avg-edges-removed\"]/(d[\"avg-num-edges\"] - 7) }))]","pinned":false,"hidden":false,"mode":"ojs"},{"id":2634,"value":"$34","pinned":false,"hidden":false,"mode":"ojs"},{"id":2545,"value":"$35","pinned":false,"hidden":false,"mode":"md"},{"id":962,"value":"simEmpiricalHeader = md`### Simulated Data Using Empirical Distributions for Drawing Distance`","pinned":false,"hidden":false,"mode":"ojs"},{"id":883,"value":"lanlHistHeader = md `#### LANL TN93 Distance Histogram`","pinned":false,"hidden":false,"mode":"ojs"},{"id":899,"value":"$36","pinned":false,"hidden":false,"mode":"ojs"},{"id":1105,"value":"LANLCaption = md`\n\nWhile the former simulations are useful for characterizing edge filtering behavior in a controlled setting, genetic distances between sequences in a transmission chain are not uniform. In order to better represent distances within a transmission chain, we fit a beta distribution over a dataset of tn93 distances calculated using the LANL HIV sequence database with a threshold of 1.5%. The beta distribution was chosen because it models the behavior of random variables limited to intervals of finite length, in this case [0, 0.015]. \n\n**α** was inferred to be **${text_float_format(2.6711811991141117)}** and **β** inferred to be **${text_float_format(0.9048171772138711)}**\n\n### Averages from 92 repetitions with chain lengths of 8\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":985,"value":"edgeDistTable = table(_.map(edge_dist, d => { \n let t = _.pick(d, ['num_nodes', \"num_edges\",\"avg-edges-removed\", \"avg-spurious\",\"avg-wrongfully-purged\", \"avg-cycles-removed\", \"avg-cycle-spurious\",\"avg-cycle-wrongfully-purged\"]); \n //t[\"avg_total_edges\"] = d[\"avg-num-edges\"] + [\"avg-edges-removed\"];\n t[\"avg-edges-removed\"] = text_float_format(t[\"avg-edges-removed\"]); \n t[\"avg-spurious\"] = text_float_format(t[\"avg-spurious\"]); \n t[\"avg-wrongfully-purged\"] = text_float_format(t[\"avg-wrongfully-purged\"]);\n t[\"avg-cycles-removed\"] = text_float_format(t[\"avg-cycles-removed\"]); \n t[\"avg-cycle-spurious\"] = text_float_format(t[\"avg-cycle-spurious\"]); \n t[\"avg-cycle-wrongfully-purged\"] = text_float_format(t[\"avg-cycle-wrongfully-purged\"]); \n t[\"FNR\"] = text_float_format(d[\"avg-spurious\"]/(d[\"avg-spurious\"] + d[\"avg-edges-removed\"]))\n t['FNR-cycle'] = text_float_format(d[\"avg-cycle-spurious\"]/(d[\"avg-cycle-spurious\"] + d[\"avg-cycles-removed\"]))\n return t\n })\n )","pinned":false,"hidden":false,"mode":"ojs"},{"id":2229,"value":"averageNumLinks = _.sum(_.map(edge_dist[\"8_edge_report.json\"].reps, d=> d[\"filter-report\"].num_edges))/_.map(edge_dist[\"8_edge_report.json\"].reps, d=> d[\"filter-report\"].num_edges).length","pinned":false,"hidden":false,"mode":"ojs"},{"id":2248,"value":"properlyReconstructedChain = _.sum(_.map(edge_dist[\"8_edge_report.json\"].reps, d=> d[\"filter-report\"].meets_edge_count_expectations && d[\"filter-report\"].num_edges == d[\"filter-report\"].true_negatives.length))/edge_dist[\"8_edge_report.json\"].reps.length","pinned":false,"hidden":false,"mode":"ojs"},{"id":2260,"value":"properlyReconstructedChainCycles = _.sum(_.map(edge_dist[\"8_edge_report.json\"].reps, d=> d[\"cycle-report\"].meets_edge_count_expectations && d[\"cycle-report\"].num_edges == d[\"cycle-report\"].true_negatives.length))/edge_dist[\"8_edge_report.json\"].reps.length","pinned":false,"hidden":false,"mode":"ojs"},{"id":2280,"value":"$37","pinned":false,"hidden":false,"mode":"ojs"},{"id":2303,"value":"FAVITESHeader = md`## Simulated FAVITES Results\n\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":2339,"value":"FAVITESSubHeader = md`#### TODO : How dense were the clusters? Cluster distribution? What fraction of false edges were removed? \n\nIf N1‹‹N2 , what can you say about the network density, that is the total number of links over the maximum number of links, Lmax?`","pinned":false,"hidden":false,"mode":"ojs"},{"id":2356,"value":"FAVITESTable = table(_.map(items, item => {\n let t = {};\n t['Simulation'] = item[0];\n t['〈C〉'] = floatFmt(item[1]['average_clustering_coefficient']);\n t['CΔ'] = floatFmt(item[1]['transitivity']);\n t['〈C〉w/ Filtering'] = floatFmt(item[2]['average_clustering_coefficient']);\n t['CΔ w/ Filtering'] = floatFmt(item[2]['transitivity']);\n return t\n}))","pinned":false,"hidden":false,"mode":"ojs"},{"id":2353,"value":"items = d3.json('https://data.hyphy.org/web/edge-filtering/items.json')","pinned":false,"hidden":false,"mode":"ojs"},{"id":2393,"value":"FAVITESDataPlot = Plot.dot(filteredDataResults, {x: \"nodes\", y: \"average_clustering_coefficient\", stroke: \"edges\"}).plot()","pinned":false,"hidden":false,"mode":"ojs"},{"id":2388,"value":"filteredDataResults = _.map(realDataResults, d=> d[1])","pinned":false,"hidden":false,"mode":"ojs"},{"id":2381,"value":"realDataResults = d3.json('https://data.hyphy.org/web/edge-filtering/all_results.json')","pinned":false,"hidden":false,"mode":"ojs"},{"id":1686,"value":"DSPSHeader = md`## Simulated DSPS Results\n\nDSPS-HIV had an FNR of **${d3.format(\".3f\")(dsps_FNR)}**, this means that **20%** of spurious edges inferred were removed with edge filtering turned on.\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1954,"value":"$38","pinned":false,"hidden":false,"mode":"ojs"},{"id":2033,"value":"degreeDistBIC = md`### Degree distribution BIC`","pinned":false,"hidden":false,"mode":"ojs"},{"id":2018,"value":"DSPSTable = table(dsps_tables)","pinned":false,"hidden":false,"mode":"ojs"},{"id":968,"value":"empiricalResultsHeader = md`## Empirical Results`","pinned":false,"hidden":false,"mode":"ojs"},{"id":2663,"value":"### AUTO-TUNE Datasets Report","pinned":false,"hidden":false,"mode":"md"},{"id":2686,"value":"viewof datasetIDKey = DOM.select([\"22739398\",\"30063225\",\"32500089\",\"32693608\",\"33281803\",\"33901684\",\"34452506\"])","pinned":false,"hidden":false,"mode":"ojs"},{"id":2665,"value":"$39","pinned":false,"hidden":false,"mode":"ojs"},{"id":2682,"value":"table(autotuneDataSetTables)","pinned":false,"hidden":false,"mode":"ojs"},{"id":3058,"value":"#### Distribution of Bayesian Information Criterion (BIC) Across Different Filtering Strategies. ","pinned":false,"hidden":false,"mode":"md"},{"id":2977,"value":"Plot.plot({\n width: 1024,\n x: {\n axis: null,\n label: null // Remove the x-axis label\n },\n fx: {\n label: null\n },\n y: {\n tickFormat: \"s\",\n grid: true,\n label: \"BIC\"\n },\n color: {\n scheme: \"spectral\",\n legend: true\n },\n marks: [\n Plot.barY(bics, {\n x: \"type\",\n y: \"bic\",\n fill: \"type\",\n fx: \"id\",\n sort: {x: null, color: null, fx: {value: \"-y\", reduce: \"sum\"}}\n }),\n Plot.ruleY([0])\n ]\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":3049,"value":"This bar chart visualizes the BIC values for various filtering approaches applied to genomic data. Categories include No Filtering, C3 Edge Filtering, C3,4 Edge Filtering, and their default counterparts. Each bar represents the BIC value associated with a specific PubMed ID and filtering type, showing the comparative effectiveness of these strategies in degree distribution model fitting.","pinned":false,"hidden":false,"mode":"md"},{"id":2979,"value":"$3a","pinned":false,"hidden":false,"mode":"ojs"},{"id":3055,"value":"#### Comparative Analysis of Mean Differences Between Filtering Methods. ","pinned":false,"hidden":false,"mode":"md"},{"id":3039,"value":"Plot.plot({\n width: 1024,\n marks: [\n Plot.barY(meanDiffs, {x: \"category\", y: \"value\", fill: \"category\"}),\n Plot.ruleY([0])\n ],\n x: {\n label: \"Difference Category\"\n },\n y: {\n label: \"Mean Difference\"\n },\n color: {\n legend: true // if you want a legend to differentiate the categories by color\n }\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":3052,"value":"This bar chart displays the mean differences in Bayesian Information Criterion (BIC) values among various filtering categories. ","pinned":false,"hidden":false,"mode":"md"},{"id":420,"value":"NCCReportHeader = md`### NCC Report`","pinned":false,"hidden":false,"mode":"ojs"},{"id":444,"value":"edgeRemovalCaption = md`#### Edge Removal\nWithout edge filtering, HIV-TRACE inferred **${ncc_no_filter.trace_results[\"Network Summary\"][\"Edges\"]}** edges. With C₃ edge filtering enabled, HIV-TRACE inferred **${ncc_filter.trace_results[\"Network Summary\"][\"Edges\"]}**, a difference of **${ncc_filter_edges_removed}** edges. When applying edge filtering with C_3,4 inspection, an additional **${ncc_additional_edges_removed}** edges were removed from the network.\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":481,"value":"$3b","pinned":false,"hidden":false,"mode":"ojs"},{"id":579,"value":"nccEdgeRemovalPlotCaption = md`**Figure 1** Degree distributions of nodes with no filtering, edge filtering, and edge filtering with cycles. Using BIC to determine model fit, the fit progressively improved with edge filtering using cycles.`","pinned":false,"hidden":false,"mode":"ojs"},{"id":607,"value":"nccEdgeRemovalTable = table(tables)","pinned":false,"hidden":false,"mode":"ojs"},{"id":587,"value":"md`**Table 1**`","pinned":false,"hidden":false,"mode":"ojs"},{"id":647,"value":"lanlReportHeader = md`### LANL Report\n\nThe LANL reference dataset, generated May 9th, 2019, included 139 060 HIV-1 group M sequences.\nWithout edge filtering, HIV-TRACE inferred **${lanl_edge_report[\"total\"]}** edges. With C₃ edge filtering enabled, HIV-TRACE inferred **${lanl_edge_report[\"total\"] - lanl_edge_report[\"filter_edges_removed\"]}**, a difference of **${lanl_edge_report[\"filter_edges_removed\"]}** edges. When applying edge filtering with C_3,4 inspection, an additional **${lanl_edge_report[\"cycles_edges_removed\"] - lanl_edge_report[\"filter_edges_removed\"]}** edges were removed from the network\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":840,"value":"$3c","pinned":false,"hidden":false,"mode":"ojs"},{"id":851,"value":"lanlTable = table(lanl_tables)","pinned":false,"hidden":false,"mode":"ojs"},{"id":1678,"value":"faq = md`## FAQ\n\n### Why can't we just remove the longest length in the triangle?\n\nThe relationship between sequence distances and time is not absolute. Substitution models are employed to take into consideration that a mutation at particular sites within an individual occur probabilistically. Since the TN93 substition model is employed to infer evolutionary distance between samples, a significance test is best suited to determine whether an edge should be removed.\n\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":2413,"value":"md`#### Can we use metadata, such as location, to determine if edge filtering is effective? `","pinned":false,"hidden":false,"mode":"ojs"},{"id":1750,"value":"md` ### Associated variables with simple example.`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1526,"value":"chain_nodes = [\n {\"id\": \"0\", \"group\": 1},\n {\"id\": \"1\", \"group\": 1},\n {\"id\": \"2\", \"group\": 1},\n {\"id\": \"3\", \"group\": 1},\n {\"id\": \"4\", \"group\": 1},\n {\"id\": \"5\", \"group\": 1},\n {\"id\": \"6\", \"group\": 1},\n {\"id\": \"7\", \"group\": 1},\n ]","pinned":false,"hidden":false,"mode":"ojs"},{"id":1532,"value":"all_chain_links = _.filter(_.flatten(_.map(chain_nodes, d => { return _.map(chain_nodes, g => { if(d.id != g.id && g.id > d.id) return { \"source\": d.id, \"target\": g.id, \"value\": 10, \"genetic_distance\" : (Number(g.id) - Number(d.id)) * mutation_rate}})})))","pinned":false,"hidden":false,"mode":"ojs"},{"id":1570,"value":"chain_links = _.filter(all_chain_links, d => d.genetic_distance <= threshold)","pinned":false,"hidden":false,"mode":"ojs"},{"id":1648,"value":"targets = _.zipWith(_.map(_.groupBy(chain_links, (d) => d.target ), d => d.length), _.map(_.groupBy(chain_links, (d) => d.source ), d => d.length), (a,b) => a+b)","pinned":false,"hidden":false,"mode":"ojs"},{"id":1807,"value":"distribution = {\nvar t2 = _.clone(_.mapValues(_.groupBy(chain_links, (d) => d.target ), d => d.length))\nvar t3 = _.clone(_.mapValues(_.groupBy(chain_links, (d) => d.source ), d => d.length))\nvar t4 = _.range(8);\nreturn _.map(t4, i => { \n return (t2[i] || 0) + (t3[i] || 0)\n })\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":6,"value":"md`# Dependencies`","pinned":false,"hidden":false,"mode":"ojs"},{"id":12,"value":"_ = require('lodash')","pinned":false,"hidden":false,"mode":"ojs"},{"id":2899,"value":"function IQR(data) {\n const sortedData = _.sortBy(data);\n const q1 = _.nth(sortedData, Math.floor(sortedData.length / 4));\n const q3 = _.nth(sortedData, Math.ceil(sortedData.length * (3 / 4)) - 1);\n return q3 - q1;\n}\n","pinned":false,"hidden":false,"mode":"ojs"},{"id":8,"value":"d3 = require('d3')","pinned":false,"hidden":false,"mode":"ojs"},{"id":3081,"value":"jStat = require('jStat');","pinned":false,"hidden":false,"mode":"ojs"},{"id":2726,"value":"moment = require('moment')","pinned":false,"hidden":false,"mode":"ojs"},{"id":2370,"value":"floatFmt = d3.format(\".2f\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":1482,"value":"d3Regression = require(\"d3-regression@1.2.3\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":10,"value":"plotly=require(\"https://cdn.plot.ly/plotly-latest.min.js\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":145,"value":"import { table } from \"@gampleman/table\"","pinned":false,"hidden":false,"mode":"ojs"},{"id":1605,"value":"import {slider} from \"@jashkenas/inputs\"","pinned":false,"hidden":false,"mode":"ojs"},{"id":4,"value":"reports = d3.json(\"https://data.hyphy.org/web/edge-filtering/2024-all-report.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":2524,"value":"edgeFilterHeaders = { return {\n columns: [\n 'internal_length',\n \"tip_length\",\n \"avg-num-edges\",\n \"avg-edges-removed\",\n \"avg-true-positives\",\n \"avg-false-positives\",\n \"avg-true-negatives\",\n \"avg-false-negatives\",\n //\"precision\",\n //\"recall\",\n //\"specificity\",\n //\"f1Score\",\n //\"accuracy\",\n \"score\"\n ],\n header: {\n 'internal_length': \"Inter-host Evo rate\",\n \"tip_length\": \"Intra-host Evo rate\",\n \"avg-num-edges\": \"Avg. Total Edges\",\n \"avg-edges-removed\": \"Avg. Edges Removed\",\n \"avg-true-positives\": \"True Positives\",\n \"avg-false-positives\": \"False Positives\",\n \"avg-true-negatives\": \"True Negatives\",\n \"avg-false-negatives\": \"False Negatives\",\n //\"precision\": \"Precision\",\n //\"recall\": \"Recall\",\n //\"specificity\": \"Specificity\",\n //\"f1Score\": \"F1 Score\",\n //\"accuracy\": \"Accuracy\",\n \"score\":\"Score\"\n }\n}}\n","pinned":false,"hidden":false,"mode":"ojs"},{"id":2576,"value":"heatMapResults = _.mapValues(reports, tableMapper)","pinned":false,"hidden":false,"mode":"ojs"},{"id":123,"value":"// update reports to have number seqs, reps, internal_length, and tip_length in object\n_.each(reports, (d,k) => { \n\n var tok = _.split(k, '_'); \n reports[k]['seqs'] = tok[0]; \n reports[k]['reps'] = tok[1]; \n reports[k]['internal_length'] = tok[2]; \n reports[k]['tip_length'] = tok[3];\n\n reports[k]['FNR'] = d[\"avg-spurious\"]/(d[\"avg-spurious\"] + d[\"avg-edges-removed\"])\n reports[k]['FNR-cycle'] = d[\"avg-cycle-spurious\"]/(d[\"avg-cycle-spurious\"] + d[\"avg-cycles-removed\"])\n \n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":1130,"value":"key_labels = edgeFilterHeaders.header","pinned":false,"hidden":false,"mode":"ojs"},{"id":238,"value":"const_tip_lengths = {\n let hi = _.groupBy(_.keys(heatMapResults), (k) => { return _.split(k, '_')[3]})\n return _.mapValues(hi, v => {return _.map(v, w => {return heatMapResults[w]})})\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":425,"value":"ncc_no_filter = d3.json(\"https://data.hyphy.org/web/edge-filtering/edge/2019-05-09-kp-net-build/KPNET.nofilter.results.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":431,"value":"ncc_filter = d3.json(\"https://data.hyphy.org/web/edge-filtering/edge/2019-05-09-kp-net-build/KPNET.results.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":436,"value":"ncc_cycle_filter = d3.json(\"https://data.hyphy.org/web/edge-filtering/edge/2019-05-09-kp-net-build/KPNET.cycle.results.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":774,"value":"ncc_filter_edges_removed = ncc_no_filter.trace_results[\"Network Summary\"][\"Edges\"] - ncc_filter.trace_results[\"Network Summary\"][\"Edges\"]","pinned":false,"hidden":false,"mode":"ojs"},{"id":452,"value":"ncc_cycle_edges_removed = ncc_no_filter.trace_results[\"Network Summary\"][\"Edges\"] - ncc_cycle_filter.trace_results[\"Network Summary\"][\"Edges\"]","pinned":false,"hidden":false,"mode":"ojs"},{"id":454,"value":"ncc_additional_edges_removed = ncc_filter.trace_results[\"Network Summary\"][\"Edges\"] - ncc_cycle_filter.trace_results[\"Network Summary\"][\"Edges\"]","pinned":false,"hidden":false,"mode":"ojs"},{"id":893,"value":"ncc_distances = d3.csv(\"https://data.hyphy.org/web/edge-filtering/edge/2019-05-09-kp-net-build/kp-aligned.fas_user.tn93output.csv\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":940,"value":"lanl_distances = d3.csv(\"https://data.hyphy.org/web/edge-filtering/edge/2019-05-11-lanl-build/raw_reference.fas_user.tn93output.csv\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":819,"value":"lanl_cycle_filter = d3.json(\"https://data.hyphy.org/web/edge-filtering/edge/2019-05-11-lanl-build/cycle.degrees.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":828,"value":"lanl_filter = d3.json(\"https://data.hyphy.org/web/edge-filtering/edge/2019-05-11-lanl-build/results.degrees.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":834,"value":"lanl_no_filter = d3.json(\"https://data.hyphy.org/web/edge-filtering/edge/2019-05-11-lanl-build/nofilter.degrees.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":837,"value":"lanl_edge_report = d3.json(\"https://data.hyphy.org/web/edge-filtering/edge/2019-05-11-lanl-build/edge_report.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":3364,"value":"(ncc_no_filter[\"trace_results\"][\"Edges\"].length - ncc_filter[\"trace_results\"][\"Edges\"].length) / ncc_no_filter[\"trace_results\"][\"Edges\"].length","pinned":false,"hidden":false,"mode":"ojs"},{"id":2800,"value":"function findDifferences(array1, array2) {\n // Convert sequences array to a string for comparison\n const stringifySequences = obj => JSON.stringify(obj.sequences.sort());\n\n // Find objects in array1 not in array2\n const diff1 = array1.filter(obj1 => \n !array2.some(obj2 => stringifySequences(obj1) === stringifySequences(obj2)));\n\n // Find objects in array2 not in array1\n const diff2 = array2.filter(obj2 => \n !array1.some(obj1 => stringifySequences(obj2) === stringifySequences(obj1)));\n\n return { uniqueInFirst: diff1, uniqueInSecond: diff2 };\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":3371,"value":"242534 / 1794851","pinned":false,"hidden":false,"mode":"ojs"},{"id":975,"value":"edge_dist = d3.json(\"https://data.hyphy.org/web/edge-filtering/edge/2019-05-12-hyphy-dist-build/all-report.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":614,"value":"text_float_format = d3.format (\".3f\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":635,"value":"interval_format = s => `[${_.map(s, text_float_format).join(',')}]`;","pinned":false,"hidden":false,"mode":"ojs"},{"id":591,"value":"tables = {\n ncc_no_filter[\"trace_results\"][\"Degrees\"][\"Type\"] = \"No Filtering\"\n ncc_filter[\"trace_results\"][\"Degrees\"][\"Type\"] = \"C₃ Edge Filtering\"\n ncc_cycle_filter[\"trace_results\"][\"Degrees\"][\"Type\"] = \"C_3,4 Edge Filtering\"\n var tables = [ncc_no_filter[\"trace_results\"][\"Degrees\"], ncc_filter[\"trace_results\"][\"Degrees\"], ncc_cycle_filter[\"trace_results\"][\"Degrees\"]]\n \n tables = _.map(tables, t => _.pick(t, [\"Type\", \"BIC\", \"rho\", \"rho CI\"]))\n \n _.each(tables, t => {\n t[\"BIC\"] = text_float_format(t[\"BIC\"]); \n t[\"ρ\"] = text_float_format(t[\"rho\"]); \n t[\"ρ CI\"] = interval_format(t[\"rho CI\"]); \n delete t[\"rho\"]\n delete t[\"rho CI\"]\n })\n \n return tables\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":853,"value":"lanl_tables = {\n lanl_no_filter[\"Type\"] = \"No Filtering\"\n lanl_filter[\"Type\"] = \"C₃ Edge Filtering\"\n lanl_cycle_filter[\"Type\"] = \"C_3,4 Edge Filtering\"\n var tables = [lanl_no_filter, lanl_filter, lanl_cycle_filter]\n \n tables = _.map(tables, t => _.pick(t, [\"Type\", \"BIC\", \"rho\", \"rho CI\"]))\n \n _.each(tables, t => {\n t[\"BIC\"] = text_float_format(t[\"BIC\"]); \n t[\"ρ\"] = text_float_format(t[\"rho\"]); \n t[\"ρ CI\"] = interval_format(t[\"rho CI\"]); \n delete t[\"rho\"]\n delete t[\"rho CI\"]\n })\n \n return tables\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":1061,"value":"tip_length_label = \"Sampling Distance sub/site\"","pinned":false,"hidden":false,"mode":"ojs"},{"id":1064,"value":"internal_length_label = \"Transmission Distance sub/site\"","pinned":false,"hidden":false,"mode":"ojs"},{"id":1269,"value":" distance = d3.text(\"https://raw.githubusercontent.com/veg/edge-filtering/master/figures/distance.svg\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":1333,"value":"drag = simulation => {\n \n function dragstarted(d) {\n if (!d3.event.active) simulation.alphaTarget(1.3).restart();\n d.fx = d.x;\n d.fy = d.y;\n }\n \n function dragged(d) {\n d.fx = d3.event.x;\n d.fy = d3.event.y;\n }\n \n function dragended(d) {\n if (!d3.event.active) simulation.alphaTarget(0);\n d.fx = null;\n d.fy = null;\n }\n \n return d3.drag()\n .on(\"start\", dragstarted)\n .on(\"drag\", dragged)\n .on(\"end\", dragended);\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":2106,"value":"layout = ({\n height: 500,\n width: 900,\n margin: { top: 20, bottom: 20, left: 20, rigth: 20 }\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":2101,"value":"subdag = await d3\n .xml(\n 'https://data.hyphy.org/web/edge-filtering/edge/two_simluation_subdag.svg'\n )\n .then(d => d.getElementsByTagName(\"svg\")[0])","pinned":false,"hidden":false,"mode":"ojs"},{"id":1387,"value":"$3d","pinned":false,"hidden":false,"mode":"ojs"},{"id":1330,"value":"$3e","pinned":false,"hidden":false,"mode":"ojs"},{"id":1928,"value":"md`### DSPS`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1933,"value":"dsps_report_raw = d3.json(\"https://data.hyphy.org/web/edge-filtering/edge/2019-09-30-dsps-build/all_edge_report.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":1971,"value":"dsps_report = dsps_report_raw[\"./rsrc/1/1_0.8/phylo_1_pol.fasta\"]","pinned":false,"hidden":false,"mode":"ojs"},{"id":1970,"value":"dsps_FNR = dsps_report[\"filter-report\"][\"spurious_edges\"].length/(dsps_report[\"filter-report\"][\"spurious_edges\"].length + dsps_report[\"filter-report\"][\"num_edges_removed\"])","pinned":false,"hidden":false,"mode":"ojs"},{"id":1939,"value":"dsps_no_filter = d3.json(\"https://data.hyphy.org/web/edge-filtering/edge/2019-09-30-dsps-build/phylo_1_pol.fasta_nodes.nofilter.results.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":1947,"value":"dsps_filter = d3.json(\"https://data.hyphy.org/web/edge-filtering/edge/2019-09-30-dsps-build/phylo_1_pol.fasta_nodes.results.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":1950,"value":"dsps_cycle_filter = d3.json(\"https://data.hyphy.org/web/edge-filtering/edge/2019-09-30-dsps-build/phylo_1_pol.fasta_nodes.cycle.results.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":2037,"value":"dsps_tables = {\n dsps_no_filter[\"trace_results\"][\"Degrees\"][\"Type\"] = \"No Filtering\"\n dsps_filter[\"trace_results\"][\"Degrees\"][\"Type\"] = \"C₃ Edge Filtering\"\n dsps_cycle_filter[\"trace_results\"][\"Degrees\"][\"Type\"] = \"C_3,4 Edge Filtering\"\n var tables = [dsps_no_filter[\"trace_results\"][\"Degrees\"], dsps_filter[\"trace_results\"][\"Degrees\"], dsps_cycle_filter[\"trace_results\"][\"Degrees\"]]\n \n tables = _.map(tables, t => _.pick(t, [\"Type\", \"BIC\", \"rho\", \"rho CI\"]))\n \n _.each(tables, t => {\n t[\"BIC\"] = text_float_format(t[\"BIC\"]); \n t[\"ρ\"] = text_float_format(t[\"rho\"]); \n t[\"ρ CI\"] = interval_format(t[\"rho CI\"]); \n delete t[\"rho\"]\n delete t[\"rho CI\"]\n })\n \n return tables\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":2268,"value":"md`### FAVITES`","pinned":false,"hidden":false,"mode":"ojs"},{"id":2271,"value":"favites_report_raw = d3.json(\"https://data.hyphy.org/web/01.sequences.aln_edge_report.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":2276,"value":"favites_report = favites_report_raw","pinned":false,"hidden":false,"mode":"ojs"},{"id":2315,"value":"favites_FNR = favites_report[\"filter-report\"][\"spurious_edges\"].length/(favites_report[\"filter-report\"][\"spurious_edges\"].length + favites_report[\"filter-report\"][\"num_edges_removed\"])","pinned":false,"hidden":false,"mode":"ojs"},{"id":2320,"value":"favitesPercentRemoved = favites_report[\"filter-report\"][\"num_edges_removed\"]/(favites_report[\"filter-report\"][\"spurious_edges\"].length + favites_report[\"filter-report\"][\"num_edges_removed\"])","pinned":false,"hidden":false,"mode":"ojs"},{"id":2287,"value":"_.filter(favites_report[\"filter-report\"][\"true_negatives\"], d => parseInt(_.split(d[0], '|')[2]) <= 0.015 && parseInt(_.split(d[1], '|')[2]) <= 0.015)","pinned":true,"hidden":false,"mode":"ojs"},{"id":2651,"value":"### AUTO-TUNE DataSets","pinned":false,"hidden":false,"mode":"md"},{"id":2655,"value":"autotune_datasets_raw = d3.json(\"https://data.hyphy.org/web/edge-filtering/autotune-datasets/all_edge_report.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":3012,"value":"autotune_datasets = {\n \n let autotune_dataset_annotated = _.pick(autotune_datasets_raw, [\"22739398\",\"30063225\",\"32500089\",\"32693608\",\"33281803\",\"33901684\",\"34452506\"]);\n\n autotune_dataset_annotated[\"22739398\"][\"author\"] = \"Dennis et al. (2012)\";\n autotune_dataset_annotated[\"30063225\"][\"author\"] = \"Rhee et al. (2019)\";\n autotune_dataset_annotated[\"32500089\"][\"author\"] = \"Yan et al. (2020)\";\n autotune_dataset_annotated[\"32693608\"][\"author\"] = \"Bbosa et al. (2020)\";\n autotune_dataset_annotated[\"33281803\"][\"author\"] = \"Liu et al. (2020)\";\n autotune_dataset_annotated[\"33901684\"][\"author\"] = \"Yan et al. (2021)\";\n autotune_dataset_annotated[\"34452506\"][\"author\"] = \"Brenner et al. (2021)\";\n\n return autotune_dataset_annotated;\n\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":3021,"value":"$3f","pinned":false,"hidden":false,"mode":"ojs"},{"id":3031,"value":"meanDiffs = {\n const meanValues = _.reduce(autotuneBiccDiffs, (acc, diffs) => {\n _.each(diffs, (value, key) => {\n if (!acc[key]) {\n acc[key] = 0;\n }\n acc[key] += value;\n });\n return acc;\n}, {});\n\n// Divide each accumulated value by the length of the array to get means\nconst meanDiffs = _.mapValues(meanValues, total => total / autotuneBiccDiffs.length);\nconst data = [\n {category: \"C3 Filter Difference\", value: meanDiffs.filterDiff},\n {category: \"C3,4 Filter Difference\", value: meanDiffs.cycleDiff},\n {category: \"C3-C3,4 Filter Difference\", value: meanDiffs.filterCycleDiff},\n {category: \"1.5% C3 Difference\", value: meanDiffs.defaultFilterDiff},\n {category: \"1.5% C3,4 Difference\", value: meanDiffs.defaultCycleDiff},\n {category: \"1.5% C3-C3,4 Difference\", value: meanDiffs.defaultFilterCycleDiff}\n];\nreturn data;\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":3146,"value":"dwhsRaw = d3.json(\"https://data.hyphy.org/web/edge-filtering/autotune-datasets/alldwh.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":3150,"value":"dwhs = Object.entries(dwhsRaw).map(([key, value]) => {\n let type;\n if (key.includes('cycle')) {\n type = 'cycle';\n } else if (key.includes('nofilter')) {\n type = 'nofilter';\n } else {\n type = 'filtered';\n }\n \n return {\n id: key.split('_')[0],\n type: type,\n ...value,\n };\n});","pinned":false,"hidden":false,"mode":"ojs"},{"id":3159,"value":"calculateAverageDWH = (assortativity) => {\n if (!assortativity || assortativity.length === 0) return 0;\n \n const values = assortativity.map(item => parseFloat(item.DWH));\n return d3.mean(values);\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":3156,"value":"dwhs.map(entry => {\n // Create a new entry object to avoid mutating the original data\n const newEntry = { ...entry };\n let sumAverageDWH = 0;\n let countAverageDWH = 0;\n \n // Iterate over each key in the entry\n Object.keys(newEntry).forEach(key => {\n const field = newEntry[key];\n \n if (field && typeof field === 'object' && field.assortativity) {\n field.averageDWH = calculateAverageDWH(field.assortativity);\n sumAverageDWH += field.averageDWH;\n countAverageDWH += 1;\n }\n });\n newEntry.overallAverageDWH = countAverageDWH > 0 ? sumAverageDWH / countAverageDWH : 0;\n return newEntry;\n });","pinned":false,"hidden":false,"mode":"ojs"},{"id":2677,"value":"$40","pinned":false,"hidden":false,"mode":"ojs"},{"id":2703,"value":"### Infection Date Report","pinned":false,"hidden":false,"mode":"md"},{"id":2784,"value":"Plot.plot({\n y: {grid: true},\n color: {legend: true},\n marks: [\n Plot.rectY(allTimeDiffs, Plot.binX({y: \"count\"}, {x: { value: \"days\", thresholds: 50}, fill:\"type\"})),\n Plot.ruleY([0])\n ]\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":2882,"value":"Plot.plot({\n facet: {\n data: allTimeDiffs,\n x: \"type\"\n },\n marks: [\n Plot.frame(),\n Plot.dot(allTimeDiffs, Plot.binX({y: \"count\"}, {x: { value: \"days\", thresholds: 50}, fill:\"type\"}))\n ]\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":3244,"value":"### Triangle Analysis","pinned":false,"hidden":false,"mode":"md"},{"id":2957,"value":"proportion = countFilteredOlder/totalTrianglesWithFiltered","pinned":false,"hidden":false,"mode":"ojs"},{"id":2947,"value":"countFilteredOlder = triangleDayDiffs.reduce((count, triangle) => {\n return count + (triangle.filteredOlder ? 1 : 0);\n}, 0);","pinned":false,"hidden":false,"mode":"ojs"},{"id":2950,"value":"totalTrianglesWithFiltered = triangleDayDiffs.reduce((acc, triangle) => {\n // Check if any edge in the triangle is filtered\n let anyFiltered = Object.entries(triangle).some(([key, value]) => \n key.includes('KX') && Array.isArray(value) && value[1] === true\n );\n\n // Increment count of triangles where filtered edges are older, if any edge is filtered\n if (anyFiltered) {\n acc += 1; // Count this triangle towards the total with filtered edges\n }\n\n return acc;\n}, 0);","pinned":false,"hidden":false,"mode":"ojs"},{"id":2722,"value":"nodesById = germanDataset[\"filter_results\"][\"Nodes\"].reduce((acc, node) => {\n acc[node.id] = node.patient_attributes;\n return acc;\n}, {});\n","pinned":false,"hidden":false,"mode":"ojs"},{"id":2775,"value":"getDiffs = (edge, type) => {\n // Extract the first and second node ids from the current edge\n let nodeId1 = edge[\"sequences\"][0];\n let nodeId2 = edge[\"sequences\"][1];\n \n // Get the infection dates for both nodes\n let infectionDate1 = nodesById[nodeId1][\"InfectionDate\"];\n let infectionDate2 = nodesById[nodeId2][\"InfectionDate\"];\n \n // Calculate the difference in days between the two dates\n let diff = Math.abs(moment(infectionDate1, \"DD-MMM-YYYY\").diff(moment(infectionDate2, \"DD-MMM-YYYY\"), 'days'));\n \n return {days: diff, type: type, from: nodeId1, to: nodeId2}; // This will be one element in the new array\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":2888,"value":"{\n return {\n allMean: _.meanBy(allTimeDiffs, d=> d.days),\n noFilterMean: _.meanBy(_.filter(allTimeDiffs, d=> d.type == \"no-filter\"), d=> d.days),\n filterMean : _.meanBy(_.filter(allTimeDiffs, d=> d.type == \"filter\"), d=> d.days),\n cycleMean : _.meanBy(_.filter(allTimeDiffs, d=> d.type == \"cycle\"), d=> d.days),\n allIQR : IQR(_.map(allTimeDiffs, d => d.days)),\n noFilterIQR : IQR(_.map(_.filter(allTimeDiffs, d=> d.type == \"no-filter\"), d => d.days)),\n filterIQR : IQR(_.map(_.filter(allTimeDiffs, d=> d.type == \"filter\"), d => d.days)),\n cycleIQR : IQR(_.map(_.filter(allTimeDiffs, d=> d.type == \"cycle\"), d => d.days))\n }\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":2929,"value":"$41","pinned":false,"hidden":false,"mode":"ojs"},{"id":2926,"value":"timeDiffLookup = allTimeDiffs.reduce((acc, obj) => {\n const key1 = obj.from + obj.to;\n const key2 = obj.to + obj.from; // because the graph might be undirected\n acc[key1] = obj.days;\n acc[key2] = obj.days; // This assumes undirected edges have the same day difference in both directions\n return acc;\n}, {});","pinned":false,"hidden":false,"mode":"ojs"},{"id":2861,"value":"allTimeDiffs = nofilterTimeDiffs.concat(filterTimeDiffs, cycleTimeDiffs)","pinned":false,"hidden":false,"mode":"ojs"},{"id":2767,"value":"nofilterTimeDiffs = noFilterDiffs.map(d => getDiffs(d, 'no-filter'));","pinned":false,"hidden":false,"mode":"ojs"},{"id":2937,"value":"filteredEdges = {\n let filteredEdges = new Set();\n\n filterDiffs.forEach(edge => {\n const sequences = edge.sequences;\n if (sequences.length === 2) {\n // Add both direction combinations to account for undirected edges\n filteredEdges.add(sequences[0] + sequences[1]);\n filteredEdges.add(sequences[1] + sequences[0]);\n }\n });\n return filteredEdges\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":2797,"value":"filterDiffs = findDifferences(germanDataset[\"nofilter_results\"][\"Edges\"], germanDataset[\"filter_results\"][\"Edges\"])[\"uniqueInFirst\"]","pinned":false,"hidden":false,"mode":"ojs"},{"id":2809,"value":"cycleDiffs = findDifferences(germanDataset[\"filter_results\"][\"Edges\"], germanDataset[\"cycle_results\"][\"Edges\"])[\"uniqueInFirst\"]","pinned":false,"hidden":false,"mode":"ojs"},{"id":2729,"value":"filterTimeDiffs = filterDiffs.map(d => getDiffs(d, 'filter'));","pinned":false,"hidden":false,"mode":"ojs"},{"id":2772,"value":"cycleTimeDiffs = cycleDiffs.map(d => getDiffs(d, 'cycle'));","pinned":false,"hidden":false,"mode":"ojs"},{"id":2819,"value":"noFilterDiffs = findDifferences(germanDataset[\"nofilter_results\"][\"Edges\"], filterDiffs.concat(cycleDiffs))[\"uniqueInFirst\"]","pinned":false,"hidden":false,"mode":"ojs"},{"id":2756,"value":"filteredInfectionDates = germanDataset[\"filter_results\"][\"Edges\"].map(edge => {\n // Extract the first and second node ids from the current edge\n let nodeId1 = edge[\"sequences\"][0];\n let nodeId2 = edge[\"sequences\"][1];\n \n // Get the infection dates for both nodes\n let infectionDate1 = nodesById[nodeId1][\"InfectionDate\"];\n let infectionDate2 = nodesById[nodeId2][\"InfectionDate\"];\n\n let collectionDate1 = nodesById[nodeId1][\"Patient ID\"];\n let collectionDate2 = nodesById[nodeId2][\"Patient ID\"];\n \n // Calculate the difference in days between the two dates\n let diff = Math.abs(moment(infectionDate1, \"DD-MMM-YYYY\").diff(moment(infectionDate2, \"DD-MMM-YYYY\"), 'days'));\n \n return [[nodeId1, infectionDate1, collectionDate1], [nodeId2, infectionDate2, collectionDate2]]; // This will be one element in the new array\n});","pinned":false,"hidden":false,"mode":"ojs"},{"id":2920,"value":"triangles = await d3.json(\"https://data.hyphy.org/web/edge-filtering/german/triangles.json\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":2706,"value":"germanDataset = {\n let ds = await d3.json(\"https://data.hyphy.org/web/edge-filtering/german/27400403_edge_report.json\")\n unpack(ds[\"filter_results\"])\n unpack(ds[\"nofilter_results\"])\n unpack(ds[\"cycle_results\"])\n unpack(ds[\"filter_report_results\"])\n return ds\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":2709,"value":"unpack = json => {\n _.each([\"Nodes\", \"Edges\"], (key) => {\n var fields = _.keys(json[key]);\n var expanded = [];\n _.each(fields, (f, idx) => {\n var field_values = json[key][f];\n if (!_.isArray(field_values) && \"values\" in field_values) {\n //console.log ('COMPRESSED');\n var expanded_values = [];\n _.each(field_values[\"values\"], (v) => {\n expanded_values.push(field_values[\"keys\"][v]);\n });\n field_values = expanded_values;\n }\n _.each(field_values, (fv, j) => {\n if (idx == 0) {\n expanded.push({});\n }\n expanded[j][f] = fv;\n });\n });\n json[key] = expanded;\n });\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":3413,"value":"betweennessTable1 = await FileAttachment(\"betweenness-Table 1.csv\").csv({typed: \"auto\"})","pinned":false,"hidden":false,"mode":"ojs"}],"files":"$W42","stdlib":"1","titling":"auto"}