Plotting - Coverage Analysyis

Question

[Gene]

I have a big file containing information about coverage depth.

File is in format

chr_position     samp1-depth value       samp2-depth-value         samp*-depth value ... ( a lot of samples: 100 +)

I want to generate a plot which will be the most appropriate representation of my data.

This is just one position of one chromosome.

10_102762159    20  314 349 283 281 506 500 186 275 210 206 589 689 521 264 341 317 353 217 154 437 310 311 197 155 1   332 226 283 229 288 176 215 330 254 421 225 282 277 378 419 608 510 158 382 268 193 166 405 187 222 311 161 151 277 355 317 309 259 312 196 148 255 185 216 414 587 259 301 564 451 615 527 278 360 224 362 292 412 216 220 366 246 183 270 385 462 575 222 275 149 515 229 183 385 225 315 142 322 272 437 223 176 328 347 373 365 198 200 184 201 350 537 535 257 283  145    493 875 357 170 154 168 793 320 365 303 271 314 250 501 202 209 332 329 260 395 256 504 593 252 275 225 350 386 237 325 345 219 243 249 382 369 447 305 262 405 237 328 376 171 330 377 275 207 171 144 440 240 84  230 258 621 145 405 197 255 419 815 383 487 554 301 356 131 222 348 132 294 168 229 463 318 714 401 181 166 451 296 347 344 534 347 262 231 235 156 215 317 664 289 278 335 316 615 530 264 331 292 278 333 459 373 147 590 607 251 287 490 258 409 623 576 315 285 171 210 551 255 283 601 302 501 560 239 321 306 190 293 396 252 268 508 340 366 260 228 488 561 579 993 200 587 446 229 351 620 262 523 176 232 218 337 186 265 396 576 286 367 366 224 210 262 507 270 317 312 179 166 235 262 573 350 383 211 335 399 213 312 241 209 308 476 257 355 264 242 171 472 269 311 216 372 430 197 229 426 198 139 398 252 397 246 356 239 505 534 285 367 362 187 303 310 280 304 291 277 196 246 249 541 160 156 255 137 283 499 336 364 220 227 879

This is my first time doing this. What do you suggest? Do you have any similar examples?

Thank you in advance.

Answer 1

[German.M.Demidov]

Basically do it by chromosome (better excluding centromere), draw a per sample LOESS (or rolling median) curve, using coordinate as x axis and depth as y-axis, and don't forget to make it semi-transparent. In my opinion it is the best you can do to provide a general overview. However, have to say, most probably your plot will be meaningless since nobody will be able to extract a significant bit of information from it.

UPD: you may also look into https://bernatgel.github.io/karyoploter_tutorial/#Examples?platform=hootsuite this software - looks pretty cool!