Effect Size or LogFC

[ghost]

Hi Jean,

How can I calculate the effect size or LogFC. Can "mle" represent effect size?

Regards,
Poorya

[JEFworks]

Hi Poorya, Yes, lb, mle, ub represent the lower bound, maximum likelihood estimate, and upper bound of the 95% confidence interval for the expression fold change on log2 scale. Jean Fan, PhD Bioinformatics and Integrative Genomics NCI F99/K00 Post-Doctoral Fellow Zhuang Lab | Harvard University 12 Oxford St, Naito 031, Cambridge, MA 02138 web: jef.works<http://jefworks.com/> On Mar 30, 2018, at 7:17 AM, Poorya Parvizi <notifications@github.com<mailto:notifications@github.com>> wrote: Hi Jean, How can I calculate the effect size or LogFC. Can "mle" represent effect size? Regards, Poorya — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub<https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_hms-2Ddbmi_scde_issues_65&d=DwMCaQ&c=WO-RGvefibhHBZq3fL85hQ&r=2gb0vmLv11Vi98WTAqlCXyDkhi11d9lKeGWDXEU-qNw&m=aMsTaNCaszknF8rme6E1R-CsFAXVrJ4NTMtdzZA0A6U&s=EX8MYALILp7FDlcofsa1rTtIKo6lp5yHY6FEwyMgH4c&e=>, or mute the thread<https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_AIj2SKP9WrVsZORihnu86K0VbeDCJk1eks5tjhRNgaJpZM4TBlkj&d=DwMCaQ&c=WO-RGvefibhHBZq3fL85hQ&r=2gb0vmLv11Vi98WTAqlCXyDkhi11d9lKeGWDXEU-qNw&m=aMsTaNCaszknF8rme6E1R-CsFAXVrJ4NTMtdzZA0A6U&s=rYbVgAIlSxgW2NAGwChT9j6egBM65K2HkVYmq9KdiMk&e=>.

[YingMa0107]

Hi, so is there a relationship between mle and lb ub? It seems that lb and ub are not symmetric on the mle. For example, my result showed that mb = -10.909825, ub=13.789589, mle=0. Or could you please tell me is there a way to obtain the standard error of logfold change? Thanks

[JEFworks]

mle = maximum likelihood estimate of the fold change lb = lower bound estimate of the fold change ub = upper bound estimate of the fold change ce = conservative estimate of expression-fold change (equals to the min(abs(c(lb, ub))), or 0 if the confidence interval crosses 0) lb and ub do not have to be symmetric about the mle because the underlying posterior probability density function may be skewed. Hope that helps. On Aug 23, 2018, at 4:47 PM, YingMa1993 <notifications@github.com<mailto:notifications@github.com>> wrote: Hi, so is there a relationship between mle and lb ub? It seems that lb and ub are not symmetric on the mle. For example, my result showed that mb = -10.909825, ub=13.789589, mle=0. Or could you please tell me is there a way to obtain the standard error of logfold change? Thanks — You are receiving this because you commented. Reply to this email directly, view it on GitHub<https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_hms-2Ddbmi_scde_issues_65-23issuecomment-2D415565830&d=DwMCaQ&c=WO-RGvefibhHBZq3fL85hQ&r=2gb0vmLv11Vi98WTAqlCXyDkhi11d9lKeGWDXEU-qNw&m=Mc5zGJlr3xFp3IzMOMUIG9rNbklsR19U35kWxfOp9gI&s=lkU2Y1ZqOVWNUEoE9ZgN-mk8tro0PE-a69fbzI-p5Ew&e=>, or mute the thread<https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_AIj2SBX3VIkkngfpFaSiAOnwxkuA3FPsks5uTxTegaJpZM4TBlkj&d=DwMCaQ&c=WO-RGvefibhHBZq3fL85hQ&r=2gb0vmLv11Vi98WTAqlCXyDkhi11d9lKeGWDXEU-qNw&m=Mc5zGJlr3xFp3IzMOMUIG9rNbklsR19U35kWxfOp9gI&s=19IbNugWfxcbjqSEK3YWsVinD7om1kPpTZqqKmJViQg&e=>.