d3.csv修改输入数据

Question

I am using d3.tsv to parse through a file. I want to change all the zeros in one column of the data(PValue column) to the next lowest number in that column. I believe the correct way of doing this is to use the accessor function but my attempts have so far failed.

d3.tsv(filename, modifyData, function(error, data) {

    data.forEach(function(d) {
        d.NAME = d.NAME;
        d.logFC = +d.logFC;
        d.logCPM = +d.logCPM;
        d.FDR = +d.FDR;
        d.PValue = +d.PValue
    });
})

When I try to do something like the following in the accessor function modifyData, I get an error saying 'data' is undefined in the code above.

function modifyData(d){
    d.forEach(function(origData){
        origData.PValue = +origData.PValue
        pValue_array.push(origData.PValue)
    })
    var pValue_array = [] 
    for (var i = pValue_array.length-1 ; i >= 0; i--){
        if (pValue_array[i] === 0){
            pValue_array.splice(i,1);
        }
    }
    var newPzero = (arrayMin(pValue_array))
    return d;
};

arrayMin is a simple function that returns the minimum value in an array. I was planning on using this value to replace all of the 0s in the PValue column. Help is greatly appreciated!

Answer 1

You can find the min value first and then replace the 0s:

d3.tsv('data.tsv', function(error, data) {

//Option A
// smallest = d3.min(data, function(d) {return +d.PValue || Infinity; })

//Option B
var noZeroes = data.filter(function(d) { return +d.Data !== 0; });
var smallest = d3.min(noZeroes, function(d) { return d.Data; })


    data.forEach(function(d) {

        d.NAME = d.NAME;
        d.logFC = +d.logFC;
        d.logCPM = +d.logCPM;
        d.FDR = +d.FDR;

        if (+d.PValue == 0 ) {
          d.Data = +smallest;
        } else {
          d.PValue = +d.PValue 
        }
    });

console.table(data);
})

Don't forget the "+" for numeric values, otherwise JS consider it as string, and your comparation will fail.-

Answer 2

You can use d3.min to get the minimum value from your data set.

For example

d3.tsv(filename, function(error, data) {
    data.forEach(function(d) {
        d.NAME = d.NAME;
        d.logFC = +d.logFC;
        d.logCPM = +d.logCPM;
        d.FDR = +d.FDR;

        d.PValue = +(d.PValue || d3.min(data, function(d) { return d.PValue || Infinity; }));
    });
})