search

WGLab / LIQA

Long-read Isoform Quantification and Analysis
Other
37 stars 12 forks source link

Issue with DAS estimation and output #38

Closed JesusMHU closed 3 months ago

JesusMHU commented 7 months ago

Hi! I'm using liqa -task diff and I got this error:

Error in optim(start, fn.opt, method = method, control = control) : non-finite finite-difference value [4] DAS testing: Gene AT1G04950 being processed...Done! DAS testing: Gene AT2G47690 being processed...Done! DAS testing: Gene AT4G10710 being processed...Done! DAS testing: Gene AT1G08660 being processed...Error in chol.default(K) : la submatriz de orden 7 no es definida positiva Error in chol.default(K) : la submatriz de orden 7 no es definida positiva Error in optim(start, fn.opt, method = method, control = control) : valor inicial en 'vmmin' no es finito DAS testing: Gene AT2G44710 being processed...Error in optim(start, fn.opt, method = method, control = control) : valor inicial en 'vmmin' no es finito DAS testing: Gene AT2G29200 being processed...Error in optim(start, fn.opt, method = method, control = control) : non-finite finite-difference value [6] Error in optim(start, fn.opt, method = method, control = control) : non-finite finite-difference value [4] DAS testing: Gene AT4G02640 being processed...Error in optim(start, fn.opt, method = method, control = control) : valor inicial en 'vmmin' no es finito Error in optim(start, fn.opt, method = method, control = control) : valor inicial en 'vmmin' no es finito

Dont know what could be causing it. Then when it finished I checked the output and it looked like this only:

AT5G51040 1 AT5G21160 0.963366399306107 AT1G07940 0.999999985914178 AT3G27470 0.650819005674937 AT2G37620 0.0892513703148758 AT3G25520 0.209992684655096 AT3G13460 0.00173611833872689 AT1G19835 0.168765918241626 AT4G39270 0.00393607527406237 AT1G15200 0.804948525950204 AT1G09570 0.109291374080427 AT1G15750 0.320812516518486 AT1G66410 0.521182338746133 AT5G62090 0.0270380833171076 AT4G35080 0.0978038264531164 AT5G55390 0.774214007413218 AT1G08570 1 AT2G40540 0.0575474227202776 AT5G03240 1 AT2G18960 0 AT4G10590 0.0536774503331144 AT5G13980 1 AT5G43310 0.576011640598243 AT2G39770 1 AT5G19140 0.984938255511173 AT5G66310 0.900582119308603 AT3G43190 1 AT1G18660 0.137197682158356 AT2G41060 0.091576656013638

JesusMHU commented 6 months ago

Hi! Any update on this? I could really really use the help here!

agouru55 commented 4 months ago

Hi, sorry for the delay! From the error messages, it looks like there was a problem with the optimization of the EM algorithm. I would try updating to the latest version of LIQA to debug and also double check the parameters you're using for the diff step along with correct formatting of the input data (esp because you were able to process some of the genes)